Liquid AI

ROLE: STRATEGIC BUSINESS DEVELOPMENT Department: Sales Location: San Francisco, Boston Location Type: Hybrid Employment Type: Full-time ABOUT LIQUID AI Liquid AI is a 3-year-old foundation-model company spun out of MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL). We build small, efficient, and incredibly fast foundation models that can run directly on the computers all around us—phones, cars, and laptops–with no internet needed, bringing AI capabilities to 99% of global compute that exists outside of data centers today. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY WE HAVE A DIFFERENTIATED PRODUCT, EARLY ENTERPRISE TRACTION, AND GROWING CUSTOMER DEMAND. NOW WE ARE BUILDING A STRATEGIC BUSINESS DEVELOPMENT TEAM THAT CAN TURN THAT MOMENTUM INTO A REPEATABLE GO-TO-MARKET MOTION. IN THIS ROLE, YOU WILL OWN REVENUE FROM THE FIRST CONVERSATION TO THE SIGNED DEAL. YOU WILL WORK DIRECTLY WITH THE FOUNDERS AND BUSINESS LEADERSHIP. YOU WILL HELP DEFINE HOW LIQUID GOES TO MARKET: WHO WE TARGET, HOW WE POSITION, HOW WE PRICE, HOW WE RUN EVALUATIONS, AND HOW WE WIN. THIS IS NOT A ROLE FOR SOMEONE WHO WANTS TO INHERIT A MATURE PLAYBOOK. RATHER, FOR SOMEONE WHO WANTS TO BUILD ONE WHILE CLOSING MEANINGFUL, FIRST OF A KIND DEALS. THE WORK - Own the full sales cycle, end to end: prospecting, outbound, discovery, technical validation, commercial negotiation, procurement, and close. - Build and manage a high-quality pipeline across priority verticals through self-sourced outbound, your network, and qualified inbound demand. - Sell Liquid Foundation Models to technical and executive buyers, including CTOs, VPs of Engineering, Heads of AI/ML, and product leaders. - Lead complex, multi-threaded enterprise deals that involve technical evaluation, security review, legal, procurement, and executive alignment. - Partner closely with engineering during technical deep-dives, proof-of-concepts, benchmarks, and deployment conversations. - Shape pricing, packaging, and deal strategy in partnership with founders and business leadership. - Be the voice of the market internally: bring back objections, competitive intelligence, pricing feedback, and emerging use cases that should influence product and positioning. - Build the early sales system: qualification criteria, deal stages, messaging, objection handling, handoff processes, and the operating cadences the next hires will scale. DESIRED EXPERIENCE Must-have: - 8+ years of revenue-driving experience in tech B2B sales or partnerships, with a track record of consistently meeting or exceeding goals, especially revenue-oriented goals. - Recent experience selling AI/ML infrastructure, developer tools, data/ML platforms, or other technically sophisticated products. - The ability to hold credible, high-trust conversations with deeply technical buyers and executives. - A proven track record of creating pipeline yourself, not just converting pipeline created for you. - Experience in an early-stage or Series A-C environment where you helped build the motion, not just execute an established one. - Comfort selling through longer, consultative enterprise cycles with multiple technical and commercial stakeholders. - Strong judgment on where to spend time, how to prioritize accounts, and how to drive a deal forward without overplaying your hand. Nice-to-have: - Experience selling into consumer electronics, automotive, life sciences, or financial services. - Familiarity with modern outbound, pipeline, and forecasting tools. - Relationships with technical decision-makers in our target account universe. WHO YOU ARE - You're energized by running deals, and every deal matters. - You are a closer, but not a transactional one. You earn trust through curiosity, rigor, and commercial sharpness. You are energized by technical products and technical buyers. You like learning how systems work, where they break, and what makes a solution genuinely better. - You want founder proximity, real ownership, and a seat where your judgment changes outcomes. - You have built in ambiguity before and know how to create structure without slowing momentum. - You care about the craft of sales: not just winning the current deal, but understanding why it was won and how to make that repeatable. WHAT SUCCESS LOOKS LIKE (YEAR ONE) - You close meaningful net-new revenue from deals you helped source and run. - Liquid has a repeatable early enterprise sales motion, with clear qualification criteria, pricing frameworks, objection handling, and handoff processes. - Founders and leadership rely on your judgment for deal strategy, pricing, pipeline prioritization, and market insight. - We have a stronger understanding of where we win, why we win, and which segments and use cases should define the next phase of GTM scale. WHAT WE OFFER - Build our commercial engine: This role shapes how Liquid AI sells, not inheriting someone else’s playbook. The work directly determines whether our technology reaches the market. - Compensation: Competitive base salary + variable compensation with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY Our models and workflows require performance work that generic frameworks don’t solve. You’ll design and ship custom CUDA kernels, profile at the hardware level, and integrate research ideas into production code that delivers measurable speedups in real pipelines (training, post-training, and inference). Our team is small, fast-moving, and high-ownership. We're looking for someone who finds joy in memory hierarchies, tensor cores, and profiler output. While San Francisco and Boston are preferred, we are open to other locations. WHAT WE'RE LOOKING FOR We need someone who: - Works profiler-first: You use tools like Nsight Systems / Nsight Compute to find bottlenecks, validate hypotheses, and iterate until improvements show up in end-to-end benchmarks. - Bridges theory and practice: You can translate ideas from papers into implementations that are robust, testable, and performant. - Executes independently: Given an ambiguous bottleneck, you can drive from profiling to kernel/integration changes to benchmarked results to maintained ownership. - Cares about the details: Memory hierarchy, occupancy, launch configs, tensor core utilization, bandwidth vs compute limits. THE WORK - Write high-performance GPU kernels for our novel model architectures - Integrate kernels into PyTorch pipelines (custom ops, extensions, dispatch, benchmarking) - Profile and optimize training and inference workflows to eliminate bottlenecks - Build correctness tests and numerics checks - Build/maintain performance benchmarks and guardrails to prevent regressions - Collaborate closely with researchers to turn promising ideas into shipped speedups DESIRED EXPERIENCE Must-have: - Authored custom CUDA kernels (not only calling cuDNN/cuBLAS) - Strong understanding of GPU architecture and performance: memory hierarchy, warps, shared memory/register pressure, bandwidth vs compute limits - Proficiency with low-level profiling (Nsight Systems/Compute) and performance methodology - Strong C/C++ skills Nice-to-have: - CUTLASS experience and tensor core utilization strategies - Triton kernel experience and/or PyTorch custom op integration - Experience building benchmark harnesses and perf regression tests WHAT SUCCESS LOOKS LIKE (YEAR ONE) - Measurable improvement on at least one critical end-to-end pipeline (throughput and/or latency), validated by repeatable benchmarks - At least one research-driven technique shipped as a production kernel and maintained over time - Performance regressions are detectable early via benchmarks/guardrails, not discovered late WHAT WE OFFER - Unique challenges: Our architectural innovations and efficiency requirements offer unique optimization challenges. High ownership from day one. - Compensation: Competitive base salary with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID AI Liquid AI is a 3-year-old foundation-model company spun out of MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL). We build general-purpose AI systems that enable advanced AI to run anywhere–from phones, cars, and laptops to satellites and on-prem infrastructure–with no internet required. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY We are hiring a GTM generalist to help Liquid’s growing business team turn our technical model leadership and early enterprise traction into a repeatable go-to-market motion by owning a portfolio of critical business activities. You will build our outbound motion and the automation layer that enables pipeline generation to scale. You will lead deals and own customer relationships. You will help us shape product development and launch new solutions by gathering customer feedback that shapes what we build and how we position it. You will leverage your understanding of the market and customer needs to create growth initiatives and partner with marketing to execute them. How you will spend your time across these activities will shift quarter to quarter based on what the business needs, and you have both the judgement and autonomy to decide how to focus your time and energy. This is a broad, high-leverage role. You will work directly with our founders, Head of Business, Head of Solutions, and marketing and sales leads to shape how a foundation model company goes to market and then own a piece of everything that decision touches. This role will give you more surface area than almost any other at Liquid and in the industry. THE WORK - Build qualified pipeline across target accounts through cold calls, highly tailored email outreach, LinkedIn, and creative prospecting. - Identify and engage technical and business stakeholders, including engineering leaders, AI/ML teams, product leaders, and executives. - Run thoughtful early-stage qualification conversations to determine where there is real fit and where we should not spend time. - Partner closely with our sales and partnerships teams on account strategy, messaging, sequencing, and handoff quality. - Step into active deal cycles when useful, especially to capture learnings that improve future pipeline generation and qualification. - Surface market insight from the front lines: objections, language that resonates, competitor patterns, use-case trends, and signals that should shape our positioning. - Help define our qualification criteria and handoff standards so that pipeline quality stays high as we scale. - Build the outbound playbook from scratch: target account strategy, messaging frameworks, channel mix, sequencing, and operating cadence. - Maintain rigorous CRM discipline from day one so we can learn quickly and build on real data. DESIRED EXPERIENCE Must have: - 4–6 years of GTM experience in B2B software, developer tools, infrastructure, or a similarly technical product category, with at least part of that time at an early-stage company. - Evidence of range, not just tenure. For example, you’ve held multiple types of GTM roles or moved from consulting into a GTM role. - Experience being directly accountable for revenue: this includes pipeline generation, quota-carrying roles, and/or strategic BD where you drove commercial results. - Experience prospecting or selling into technical personas such as engineers, data scientists, CTOs, Heads of AI/ML, or technical founders. - Fluency with modern GTM tooling, including at least one example of a workflow or automation you built yourself. Nice-to-have: - Experience in a Series A-C or other early-stage environment. - Familiarity with AI/ML concepts or experience selling into technical infrastructure buyers. - Experience using qualification frameworks such as MEDDIC, BANT, or similar. WHO YOU ARE We’re looking for someone who is excited by the idea of helping build a go-to-market motion from the ground up. The strongest candidates will bring: - Proof of range: you enjoy and are effective across pipeline development, deal work, customer engagement, and growth campaigns. - A builder mindset: you do not need a finished playbook. In fact, helping create one is a major part of the appeal of this role. - Excellent written communication skills: you enjoy writing and have the skills and judgement to craft materials that are relevant, specific, and worth reading / responding to. - Persistence and consistency: you understand that outbound is a game of iteration, signal gathering, and sustained effort. - Curiosity and coachability: you learn quickly from calls, reply patterns, objections, and feedback from the team. - Commercial judgement: you know the difference between activity and progress, and you care about pipeline quality, not just volume. WHAT SUCCESS LOOKS LIKE (YEAR ONE) - Pipeline: you generate a meaningful share of qualified pipeline, and the automated systems you built create opportunities without constant manual effort. - Deals: you have closed revenue you sourced or carried, and the customer relationships you own are strong and expanding. - Market testing: leadership decisions about new initiatives cite customer feedback you gathered, and product and positioning have changed for the better because of it. - Growth: campaigns you scoped have been executed with marketing and produced measurable results. - The GTM playbook (targeting, messaging, qualification, campaign process) is better documented, more automated, and more repeatable than when you arrived. WHAT WE OFFER - Build our commercial engine: This role shapes how Liquid AI sells, not inheriting someone else’s playbook. The work directly determines whether our technology reaches the market. - Compensation: Competitive base salary + variable compensation with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

Liquid AI is an efficiency-first foundation model company. We build highly capable AI models designed to operate across the full range of environments—from cloud infrastructure to phones, vehicles, and other edge devices. OPPORTUNITY Liquid AI is seeking an experienced Associate General Counsel to lead our product counseling and AI regulatory work. Reporting to the Chief Legal Officer, this person will work closely with our product, research, and engineering teams as they develop and deploy new AI models and products. The role requires someone who can understand novel and unsettled legal issues, make practical decisions, and translate those decisions into things the company can actually do. The job is not simply to monitor regulatory developments or write policies. It is to determine what they mean for Liquid’s products and help the company act. CORE RESPONSIBILITIES - Own legal support for the development, release, and deployment of Liquid’s models and products. - Advise product, research, and engineering teams on AI regulation, privacy, data use, intellectual property, model safety, security, and product claims. - Lead Liquid’s response to the EU AI Act and other emerging AI laws, including the practical controls, documentation, and product decisions needed for compliance. - Advise on training and evaluation data, synthetic data, model outputs, open-source and open-weight releases, fine-tuning, and customer deployments. - Support the commercial team on product and regulatory issues arising in strategic customer and partnership agreements. REQUIRED CAPABILITIES - At least 8 years of relevant legal experience, including meaningful in-house experience advising technology product teams. - Strong background in AI regulation, privacy, data governance, intellectual property, or related technology issues. - Ability to understand how Liquid’s products work and give clear, practical, risk-based advice. - Ability to operate independently, identify issues before they become problems, and drive work through completion. - Comfort making decisions in areas where the law is new, unsettled, or simply does not provide a clean answer. - Clear, concise writing. THE RIGHT CANDIDATE - Thinks from first principles and is genuinely interested in technology. - Wants to work directly with the people building products—not review them from a distance. - Exercises judgment rather than reflexively escalating, documenting, or avoiding risk. - Is comfortable taking ownership, being decisive, and occasionally being wrong. - High integrity, optimistic, and low ego. - Healthy sense of humor.

LIQUID AI JOB DESCRIPTION Role: Member Of Technical Staff, Infrastructure Department: Research & Engineering Location: Boston Location Type: Hybrid Employment Type: Full-time ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY Our inference stack is central to everything we ship. You'll be a core part of the team responsible for the engine layer that runs our models in production and in partner environments, and for the benchmarking infrastructure we use to evaluate our own work and verify what partners bring to us. Day to day, that means working closely with research and product, but also directly with external engineering teams. WHAT WE'RE LOOKING FOR We need someone who: - Can pick up unfamiliar tools quickly and knows how to assess whether they're worth using. - Designs AI benchmarks and holds methodology to a high standard. - Cares about inference details, understands the tradeoffs, and checks what changed across the board before calling something done. - Doesn’t consider a model port finished until you can prove the outputs are correct. THE WORK - Design and build benchmark suites that cover inference performance, model quality, and knowledge evaluation across different hardware targets. - Run external partner verifications: evaluate their solutions against our benchmarks, identify gaps, and clearly deliver findings. - Port models like LFM2 onto different runtimes and frameworks, and verify correctness end-to-end. - Maintain and extend the inference engine layer built on llama.cpp, ONNX, and MLX as new model architectures emerge from research. - Make benchmark results explainable and verifiable, so internal teams and partners can trust and reproduce them independently. DESIRED EXPERIENCE Must-have: - Hands-on experience with at least one inference framework like llama.cpp, ONNX Runtime, or MLX, going beyond basic usage into internals and modification. - Experience designing and building benchmarking pipelines, including methodology, validation, and reproducibility. - Strong C++ and Python in performance-sensitive contexts. - Solid understanding of inference fundamentals: quantization, decoding strategies, memory layout, and how they interact. Nice-to-have: - Experience porting models across runtimes and verifying numerical correctness. - Prior work with external partners or clients in a technical validation or evaluation capacity. - Familiarity with edge inference targets and the constraints that come with them. WHAT SUCCESS LOOKS LIKE (YEAR ONE) - You've ported LFM2 onto multiple runtimes and platforms, you know the model inside out, and new ports take you a fraction of the time they did at the start. - You've run multiple partner verifications end-to-end and built enough context to spot weak evaluations quickly and push back with evidence. - The benchmark suite covers inference performance and model quality across the platforms we care about, and both internal teams and partners are using it as a reference. WHAT WE OFFER - Compensation: Competitive base salary with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY Liquid AI is building a solutions architecture function from scratch. You will be one of the first SAs, working directly with the Head of Solutions Architecture and across the go-to-market org to own customer engagements end-to-end. Our models are purpose-built for environments where memory, latency, and power are binding constraints - edge devices, mobile, embedded systems, and on-prem infrastructure where frontier models simply cannot run. You will work at this boundary every day. Customers range from AI-native companies to enterprise organizations exploring AI for the first time. Your job is to bridge the gap between what our models can do and what customers believe is possible, then deliver on that promise from technical validation through go-live. WHAT WE'RE LOOKING FOR We need someone who: - Technical builder: You can download a model, build a demo, and present it to a customer. You are as comfortable in a Jupyter notebook as you are in a boardroom. - Creative problem solver: You see opportunities where customers see limitations. You can take a small, efficient model and show an enterprise why it changes their cost structure or enables something they did not think was possible. - End-to-end owner: You do not draw a line between 'pre-sales' and 'post-sales.' You own the outcome from first call to go-live and beyond. - Org builder: You want to build a function, not inherit one. You will create playbooks, demo libraries, and engagement processes that scale as the team grows. - Imagination-gap closer: Enterprise buyers often cannot envision what a fine-tuned small model can do at middleware speeds. You don't just demo—you reframe what's possible on hardware they already own. THE WORK - Own customer engagements end-to-end: from qualified opportunity through technical validation, go-live, and ongoing delivery across all customer segments - Build customer-specific demos and proofs-of-concept using Liquid models (including LEAP for fine-tuning, domain adaptation, and evaluation) to drive technical wins - Lead technical discovery: map current-state customer architectures to Liquid solutions, drive competitive positioning against open-source and incumbent models, and quantify ROI for both cost-optimization and new-experience use cases - Co-own the product-field feedback loop: document friction patterns, eval failures, and capability gaps from engagements and partner with product and research to influence roadmap - Turn engagement learnings into reusable assets: reference architectures, solution primitives, demo building blocks, engagement playbooks, and vertical-specific solution patterns across Liquid's priority industries DESIRED EXPERIENCE Must-have: - Applied ML skills: hands-on experience working with ML models in customer-facing contexts (building demos, prototypes, or production integrations) - Pre-sales and post-sales experience: you have owned technical customer engagements end-to-end, not just the pitch - Strong customer-facing communication: you can run discovery, build relationships with technical and business buyers, and present to executives - Understanding of AI architectures and deployment tradeoffs: token efficiency, on-device vs. cloud, model size vs. latency, open-weight vs. proprietary Nice-to-have: - Familiarity with small or efficient model deployment (edge, on-device, latency-constrained environments) - Track record of creating thought leadership content, technical blogs, or presenting at industry events - Familiarity with efficient model deployment: quantization (INT4/INT8, GGUF, AWQ), model serving frameworks (vLLM, TensorRT-LLM, llama.cpp), and hardware-aware optimization for edge or latency-constrained environments - Experience designing and debugging model evaluations—you understand why benchmark results can diverge from production performance and know how to diagnose the root cause WHAT SUCCESS LOOKS LIKE (YEAR ONE) 1. Qualified opportunities convert to technical wins faster, with a measurable improvement in the qualified-to-win rate 2. A library of scalable demos, engagement playbooks, and customer-facing collateral exists and is actively used 3. A structured feedback loop from customer conversations to the product and model teams is established and influencing roadmap decisions WHAT WE OFFER - Build the function: You are defining how Liquid goes to market technically, with direct influence on product direction and access to the founding team. - Compensation: Competitive base salary with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY We're hiring a Product Manager to drive specific product bets from idea to product-market fit. This is a hands-on, ground-level role: you own an experiment, or a small set of them, get the product into real users' hands, and iterate toward PMF, working shoulder to shoulder with our ML, inference, and engineering teams to turn their energy into crisp, executable direction. It is not a traditional SaaS PM role: the model and the product are inseparable, our core offering is customization & fine-tuning rather than prompt-only use, and real model depth is a prerequisite. WHAT WE'RE LOOKING FOR We need someone who: - Focus-giver: You turn high-energy, unfocused ideas into tight, executable product direction that engineers can build against. - Evidence over intuition: You put products in front of real users early and let honest feedback drive decisions. - Technically fluent and AI-native: You hold your own with ML and inference engineers, know when an LFM is the right tool, and use AI & coding tools to prototype and test hypotheses yourself. - Conviction without stubbornness: You hold a clear point of view, update it in light of evidence, and can hit the ground running with light support. THE WORK - Own one or more active product experiments end-to-end and drive them toward product-market fit. - Turn vague, high-energy ideas into tight, executable product requirements that engineers can build against. - Put the product in real users' hands early, gather genuine feedback, and iterate. - Ruthlessly prioritize: decide what to build now, what to defer, and what to kill. - Work closely with ML, inference, and engineering teams, and flex across bets as priorities shift. DESIRED EXPERIENCE Must-have: - Direct, hands-on ML or applied LLM experience: you have built or trained models, you know the difference between prompting and fine-tuning in practice, and you know when an LFM is the right tool. - A strong bias for action: you use AI and coding tools to prototype and test hypotheses yourself, so your insights are higher-signal than a PM who ran deep research once. - Demonstrated ownership of product direction and prioritization, not just execution against a handed-down roadmap. - The ability to turn an ambiguous idea into a concrete, buildable plan, with light support rather than heavy coaching. Nice-to-have: - Enterprise product expertise: the focus and execution it takes to bring a new enterprise product to market and win in enterprise. - Experience at an AI/ML company or on AI-powered products. - Familiarity with edge or on-device inference, agentic harnesses, evals, or observability. WHAT SUCCESS LOOKS LIKE (YEAR ONE) 1. You own a bet end to end and take it from an ambiguous opportunity to a validated, or confidently killed, product direction backed by real user evidence. 2. The engineers you support move faster because your requirements are crisp and well-prioritized. 3. You have established a repeatable way to get products in front of users and turn their feedback into decisions. WHAT WE OFFER - Focused ownership: You drive a product bet end to end rather than owning a sliver of a large platform. - Compensation: Competitive base salary with equity in a unicorn-stage company. - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents. - Financial: 401(k) matching up to 4% of base pay. - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year.

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY Our Audio team is building frontier speech-language models that handle STT, TTS, and speech-to-speech in a single architecture. This role sits at the center of applied audio model development, working directly with the technical lead to ship production systems that run on-device under real-time constraints. You will own critical workstreams across data pipelines, evaluation systems, and customer deployments. If you want high ownership on rare technical problems in a small, elite team where your code ships, this is the role. WHAT WE'RE LOOKING FOR We need someone who: - Builds first, theorizes later: You ship working systems, not just notebooks. Production-grade code is your default, not a stretch goal. - Owns outcomes end-to-end: From data pipelines to customer deployments, you take responsibility for the full stack without waiting for someone else to handle the hard parts. - Thrives under constraints: On-device, low-latency, memory-limited systems excite you. You see constraints as design parameters, not blockers. - Ramps quickly on new territory: Gaps in specific subdomains are fine if you close them fast. You seek out feedback and stay focused on what moves the needle. THE WORK - Build and scale data pipelines for audio model training, including preprocessing, augmentation, and quality filtering at scale - Design, implement, and maintain evaluation systems that measure multimodal performance across internal and public benchmarks - Fine-tune and adapt audio models for customer-specific use cases, owning delivery from requirements through deployment - Contribute production code to the core audio repository, collaborating with infrastructure and research teams - Support experimentation under real hardware constraints, shifting between customer work and core development as priorities evolve DESIRED EXPERIENCE Must-have: - Strong programming fundamentals with demonstrated ability to write clean, maintainable, production-grade code - Experience building and shipping production ML systems beyond model training (data pipelines, evals, serving infrastructure) - Proficiency in PyTorch and familiarity with distributed training frameworks (DeepSpeed, FSDP, or similar) - Track record of collaborating effectively in shared codebases with high engineering standards Nice-to-have: - Direct experience with audio/speech models (ASR, TTS, vocoders, diarization, or speech-to-speech systems) - Experience designing and running large-scale training experiments on distributed GPU clusters - Open-source contributions that demonstrate code quality and engineering judgment WHAT SUCCESS LOOKS LIKE (YEAR ONE) - Within 6 months, you independently deliver production-ready data pipelines or evaluation systems and own at least one customer workstream end-to-end - Your PRs to the core audio repo are accepted without heavy rework, demonstrating strong judgment in system design - By year end, you operate as a second pillar to the technical lead, unblocking parallel workstreams and raising overall team velocity WHAT WE OFFER - Rare technical problems: Work on audio-to-audio frontier systems with real ownership in a team small enough that your contributions ship directly to production. - Compensation: Competitive base salary with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY Our Training Infrastructure team is building the distributed systems that power our next-generation Liquid Foundation Models. As we scale, we need to design, implement, and optimize the infrastructure that enables large-scale training. This is a high-ownership training systems role focused on runtime/performance/reliability (not a general platform/SRE role). You’ll work on a small team with fast feedback loops, building critical systems from the ground up rather than inheriting mature infrastructure. WHAT WE'RE LOOKING FOR We need someone who: - Loves distributed systems complexity: Our team builds systems that keeps long training runs stable, debugs training failures across GPU clusters, and improves performance. - Wants to build: We need builders who find satisfaction in robust, fast, reliable infrastructure. - Thrives in ambiguity: Our systems support model architectures that are still evolving. We make decisions with incomplete information and iterate quickly. - Aligns with team priorities and delivers: Our best engineers align with team priorities while pushing back with data when they see problems. THE WORK - Design and build core systems that make large training runs fast and reliable - Build scalable distributed training infrastructure for GPU clusters - Implement and tune parallelism/sharding strategies for evolving architectures - Optimize distributed efficiency (topology-aware collectives, comm/compute overlap, straggler mitigation) - Build data loading systems that eliminate I/O bottlenecks for multimodal datasets - Develop checkpointing mechanisms balancing memory constraints with recovery needs - Create monitoring, profiling, and debugging tools for training stability and performance DESIRED EXPERIENCE Must-have: - Hands-on experience building distributed training infrastructure (PyTorch Distributed DDP/FSDP, DeepSpeed ZeRO, Megatron-LM TP/PP) - Experience diagnosing performance bottlenecks and failure modes (profiling, NCCL/collectives issues, hangs, OOMs, stragglers) - Understanding of hardware accelerators and networking topologies - Experience optimizing data pipelines for ML workloads Nice-to-have: - MoE (Mixture of Experts) training experience - Large-scale distributed training (100+ GPUs) - Open-source contributions to training infrastructure projects WHAT SUCCESS LOOKS LIKE (YEAR ONE) - Training throughput has increased - Overall training efficiency/cost has improved - Training stability has improved (fewer failures, faster recovery) - Data loading bottlenecks are eliminated for multimodal workloads WHAT WE OFFER - Greenfield challenges: Build systems from scratch for novel architectures. High ownership from day one. - Compensation: Competitive base salary with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY This is a rare chance to apply frontier sequential recommendation architectures to real enterprise problems at scale. You will own applied ML work end-to-end for recommendation system workloads, adapting Liquid Foundation Models for customers who need personalization and ranking capabilities that run efficiently under production constraints. Unlike most recommendation roles that are siloed into a single product surface, this role gives you full ownership over how large-scale recommendation models are adapted, evaluated, and deployed for enterprise customers. Between engagements, you will build reusable applied tooling and workflows that accelerate future delivery. If you care about data quality at scale, user behavior modeling, and making recommendation systems actually work in enterprise production environments, this is the role. WHAT WE’RE LOOKING FOR We need someone who: - Takes ownership: Owns customer recommendation system engagements end-to-end, from requirements through delivery and evaluation. - Thinks at scale: Can reason about user interaction data, sequential modeling, feature engineering, and evaluation across large-scale production systems. - Is pragmatic: Optimizes for measurable customer outcomes (engagement, conversion, revenue lift) over theoretical novelty. - Communicates clearly: Can translate between customer business metrics and internal technical decisions, and push back when needed. THE WORK - Act as the technical owner for enterprise customer engagements involving recommendation and ranking workloads - Translate customer requirements into concrete specifications for recommendation models - Design and execute data pipelines for user interaction data, feature engineering, and training data curation at scale - Fine-tune and adapt large-scale sequential recommendation models (e.g., HSTU-style architectures) for customer-specific use cases - Design task-specific evaluations for recommendation model performance (ranking quality, latency, throughput) and interpret results - Build reusable applied tooling and workflows that accelerate future customer engagements DESIRED EXPERIENCE Must-have: - Hands-on experience building or fine-tuning recommendation models at scale (not just off-the-shelf collaborative filtering) - Experience with sequential recommendation architectures, user behavior modeling, or large-scale ranking systems - Strong intuition for data quality and evaluation design in recommendation contexts (offline metrics, A/B testing, business metric alignment) - Experience with large-scale data pipelines for user interaction data and feature engineering - Proficiency in Python and PyTorch with autonomous coding and debugging ability Nice-to-have: - Experience with transformer-based recommendation architectures (HSTU, SASRec, BERT4Rec, or similar) - Experience delivering recommendation systems to external customers with measurable business outcomes - Familiarity with serving recommendation models under latency and throughput constraints WHAT SUCCESS LOOKS LIKE (YEAR ONE) - Independently owns and delivers enterprise recommendation system engagements with minimal oversight - Is trusted by customers as the technical owner, demonstrating strong judgment on the tradeoffs between model quality, latency, and business impact - Has built reusable applied workflows or tooling that accelerate future customer engagements WHAT WE OFFER - Real ML work: You will build and adapt large-scale recommendation models for enterprise customers, working with frontier architectures like HSTU under real production constraints. - Compensation: Competitive base salary with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY This is a rare chance to own applied post-training work end-to-end for text workloads, adapting Liquid Foundation Models for some of the world’s largest enterprise customers. You will act as the technical bridge between customer requirements and model delivery. You will lead engagements from scoping through evaluation, with full ownership over how text models are adapted and shipped. Between engagements, you will build reusable applied workflows and tooling that accelerate future delivery. If you care about data quality, evaluation design, and making language models actually work in production for real customers, this is the role. WHAT WE'RE LOOKING FOR We need someone who: - Takes ownership: Owns customer post-training projects end-to-end, from requirements through delivery and evaluation. - Thinks end-to-end: Can reason across data generation, instruction tuning, alignment, and evaluation as a single system. - Is pragmatic: Optimizes for model quality and customer outcomes over publications or theory. - Communicates clearly: Can translate between customer needs and internal technical teams, and push back when needed. THE WORK - Act as the technical owner for enterprise customer post-training engagements involving text workloads - Translate customer requirements into concrete post-training specifications and workflows - Design and execute data generation, filtering, and quality assessment processes for text corpora - Run supervised fine-tuning, instruction tuning, RLHF, DPO, and other preference alignment workflows - Design task-specific evaluations for text model performance and interpret results - Build reusable applied tooling and workflows that accelerate future customer engagements DESIRED EXPERIENCE Must-have: - Hands-on experience with data generation and evaluation for LLM post-training - Experience training or fine-tuning models using SFT, instruction tuning, RLHF, DPO, or similar preference alignment methods - Strong intuition for text data quality and evaluation design - Experience with text-specific post-training workflows: chat model alignment, instruction tuning, or text data curation at scale - Proficiency with open-source ML ecosystem (Hugging Face, PyTorch) and modern model architectures Nice-to-have: - Experience delivering applied ML work to external customers with measurable outcomes - Familiarity with inference optimization frameworks (vLLM, SGLang, TensorRT) - Experience building reusable ML tooling or evaluation infrastructure WHAT SUCCESS LOOKS LIKE (YEAR ONE) - Independently owns and delivers enterprise post-training projects for text workloads with minimal oversight - Is trusted by customers as the technical owner, demonstrating strong judgment and delivery quality - Has built reusable applied workflows or tooling that accelerate future customer engagements WHAT WE OFFER - Real ML work: You will fine-tune models, generate data, and ship solutions, not configure API calls. Your work feeds directly back into our core model development. - Compensation: Competitive base salary with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY As an ML Scientist focused on post-training, you will improve the capabilities and behavior of Liquid Foundation Models for Japan and the global market. You will own the full experimental loop: identifying model weaknesses, developing data and training strategies, running controlled experiments, evaluating results, and turning successful ideas into high-quality checkpoints and reusable methods. You will work closely with Liquid's model, research, and infrastructure teams in Japan and at our US headquarters. Signals from real deployments will give you concrete, difficult problems to solve; your work will translate those signals into broader improvements in model quality, reliability, and usefulness. This is a role for a scientist who builds. Strong ideas matter, but so do careful data work, reliable implementations, and measurable improvements in working models. WHAT WE'RE LOOKING FOR We need someone who: - Thinks experimentally: You form clear hypotheses, establish strong baselines, design informative ablations, and let evidence guide the next step. - Understands models through data: You can identify gaps in training data and design curation, filtering, generation, and supervision strategies to address them. - Owns the full post-training loop: You are comfortable moving between datasets, training, evaluation, error analysis, and implementation. - Builds with rigor: Your experiments are reproducible, your conclusions are well supported, and successful research can be integrated into shared systems. - Collaborates across boundaries: You communicate clearly with globally distributed research, engineering, and applied teams. THE WORK - Design and execute post-training strategies for language and multimodal models, including supervised fine-tuning, preference optimization, reinforcement learning, and distillation. - Build and curate high-quality training data using human, synthetic, and model-generated signals, with particular attention to Japanese-language and domain-specific capabilities. - Develop evaluations that expose meaningful capability and reliability gaps across Japanese and global use cases. - Conduct systematic error analysis and use the results to improve data mixtures, objectives, training methods, and model behavior. - Run controlled experiments and ablations, interpret results, and communicate clear recommendations. - Develop reliable, scalable training and evaluation pipelines in collaboration with model infrastructure teams. - Contribute methods, tooling, datasets, and findings that accelerate post-training work across Liquid AI. DESIRED EXPERIENCE Must-have: - Hands-on experience post-training modern language or multimodal models. - Strong understanding of machine learning fundamentals and current post-training and RL methods. - Solid engineering skills and proficiency with the open-source ML ecosystem. - Experience designing and running rigorous experiments, including baselines, ablations, and systematic error analysis. - Experience building, curating, or assessing training and evaluation data at a meaningful scale. - Ability to turn research ideas into reliable implementations and measurable model improvements. - Proficiency in English, including the ability to collaborate on complex technical work with global teams. - Experience leveraging agents to amplify your own work. Nice-to-have: - Experience with preference optimization or reinforcement learning methods for foundation models. - Experience post-training multimodal models involving text, vision, or audio. - Experience developing synthetic data pipelines, reward models, verifiers, or model-based evaluations. - A track record of producing useful research artifacts, such as strong models, datasets, open-source systems, or technical reports. - Reading proficiency in Japanese. We care about demonstrated ability, scientific judgment, and the quality of your work. Degrees and publications are not required, and we encourage you to apply even if your experience does not match every item above. WHAT SUCCESS LOOKS LIKE (YEAR ONE) 1. You deliver checkpoints with clear, measurable improvements on important Japanese and global capabilities. 2. You establish evaluations and error-analysis practices that reliably identify model weaknesses and guide post-training priorities. 3. You develop data or training methods that become part of Liquid's repeatable post-training workflow. 4. You run high-quality experiments at increasing scale and turn the results into clear technical decisions. 5. Your work materially influences Liquid's model roadmap and improves systems used in real deployments. HOW WE WORK This is a primarily remote role for candidates already based in Japan. We typically meet in our Tokyo office once or twice a week. Working hours are fully flexible, with an expectation that you can collaborate effectively with customers in Japan and colleagues in other time zones. The role includes travel to the United States and to relevant conferences, as well as occasional travel within Japan. WHAT WE OFFER - High-impact model work: Improve the capabilities and behavior of foundation models used in real products and deployments. - Direct access to core teams: Collaborate closely with the researchers and engineers building Liquid's models and training systems. - Ownership in a growing market: Help define our post-training approach in Japan at a time of strong customer demand and rapid growth. - Supportive & growth-driven culture: Thrive in a collaborative, feedback-rich environment that prioritizes continuous learning, mentorship, and personal career development. - Flexibility: Primarily remote work, flexible working hours, and unlimited paid time off. - Competitive rewards: Competitive salary, equity in a unicorn-stage company, and standard benefits for employees in Japan.

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY Our Data team powers Liquid Foundation Models across pre-training, vision, audio, and emerging modalities. Public data sources are plateauing. Model performance increasingly depends on purpose-built datasets. We need ML-minded engineers who can collect, filter, and synthesize high-quality data at scale. We treat data as a research problem, not an infrastructure problem. Our engineers run experiments, design ablations, and measure how data decisions move model quality. We will match you to the team where you can grow the fastest and have the most impact: pre-training, post-training RL, vision-language, audio, or multimodal. While San Francisco and Boston are preferred, we are open to other locations. WHAT WE'RE LOOKING FOR We need someone who: - Thinks like a researcher, ships like an engineer: We need people who form hypotheses, run experiments, and measure results. Our engineers understand deep-theoretical research, and our researchers ship production systems. - Learns fast and adapts: We work across modalities that evolve weekly. We need people who pick up new domains quickly and thrive with ambiguity. - Obsesses over data quality: We believe data quality is non-negotiable. Filtering, deduplication, augmentation, and evaluation are first-class concerns for our team, not afterthoughts. - Solves problems independently: Our data engineers sit within training groups (pre-training and multimodal). We collaborate closely, but we expect ownership and self-direction. THE WORK - Build and maintain data processing, filtering, and selection pipelines at scale - Create pipelines for pretraining, midtraining, SFT, and preference optimization datasets - Design synthetic data generation systems using LLMs, structured prompting, and domain-specific generators - Design and run evaluations and ablations to measure dataset's impact on model performance - Monitor public datasets across text, vision, and audio domains - Collaborate with pre-training, vision, and audio teams on modality-specific data needs DESIRED EXPERIENCE Must-have: - Strong Python skills with the ability to quickly comprehend problems and translate them into clean, working code - Solid ML fundamentals: experience training, evaluating, and iterating on models (PyTorch preferred) - Track record of learning new technical domains quickly - 3+ years relevant experience with an M.S., or 1+ year with a Ph.D. (5+ years with a B.S.) Nice-to-have: - Experience with synthetic data generation, data curation, or ML evaluation (designing evals, benchmarking, measuring data and model quality) - Experience with LLMs, VLMs, computer vision, or audio data pipelines - Open-source contributions or publications at NeurIPS, ICML, ICLR, or CVPR WHAT SUCCESS LOOKS LIKE (YEAR ONE) - You own a critical data pipeline end-to-end for one of our modalities - You have built or improved data systems that measurably moved model performance - You have identified and integrated at least one external dataset that moved the needle WHAT WE OFFER - Impact at scale: Your pipelines directly determine model quality across all of Liquid's foundation models. - Compensation: Competitive base salary with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID LABS Research has been core to Liquid AI from the beginning. Liquid Labs gives that work a formal home; an internal research accelerator driving fundamental breakthroughs in the science of building intelligent, personalized, and adaptive machines. Our origins trace back to MIT CSAIL, where the foundational work on Liquid Neural Networks defined a new class of dynamical, efficient sequence-processing architectures. That research became the basis for Liquid Foundation Models (LFMs). Scalable, multimodal models built for real-world deployment in resource-constrained environments. At Liquid Labs, we extend that lineage - pushing forward the frontier of efficient, adaptive intelligence through both fundamental research and practical engineering. We work hand-in-hand with Liquid’s core foundation model and systems teams to translate theory into deployed capability — defining a new generation of intelligent systems that are both powerful and efficient. ABOUT THE ROLE: As a Research Engineer, you’ll join a small, high-context team exploring the limits of adaptive intelligence. You’ll design and implement novel architectures, training methods, and inference strategies to redefine what efficient AI can do. You’ll operate at the intersection of research and engineering — translating scientific ideas into working systems, publishing where it drives the field forward, and deploying where it changes what’s possible. While San Francisco and Boston are preferred, we are open to other locations in the United States. THIS ROLE IS FOR YOU IF: - Work fluently in Python and frameworks such as PyTorch, JAX, or TensorFlow - Have experience in machine learning research or production-grade ML systems - Move fast from paper to prototype — curiosity backed by precision - Care about efficiency, scalability, and elegant system design as scientific principles - Value small, deep-technical teams where impact is immediate and measurable - Have a track record of publication in tier-1 venues (NeurIPS, ICML, ICLR, CVPR, ACL, or equivalent), demonstrating original contribution and research rigor OPEN SCIENCE AND IMPACT Liquid Labs reinforces our commitment to transparent, reproducible, open research. We publish through technical reports, architectural deep dives, ablations, and model releases, advancing the broader science of efficient AI while translating breakthroughs into production-ready systems. WHY LIQUID LABS Liquid Labs is for researchers who build. Those who care about lasting impact more than publication count, but who hold themselves to the same scientific standard. We don’t chase benchmarks; we redefine them. We move fast, think deeply, and measure success by the systems that endure. There is no application deadline. We review candidates on a rolling basis.

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY This is a rare chance to sit at the intersection of frontier vision-language models and real-world deployment. You'll own applied post-training work for VLMs end-to-end for some of the world's largest enterprises, while still contributing directly to Liquid's core multimodal model development. Unlike most roles that force a trade-off between customer impact and foundational work, this role gives you both: deep ownership over how vision-language models are adapted, evaluated, and shipped, and a direct line into the evolution of Liquid's multimodal post-training stack. If you care about visual understanding, data quality, evaluation, and making VLMs actually work in production, this is a chance to shape how applied multimodal AI is done at a foundation model company. WHAT WE'RE LOOKING FOR We need someone who: - Takes ownership: Owns VLM post-training projects end-to-end, from customer requirements through delivery and evaluation. - Thinks end-to-end: Can reason across visual data curation, training, alignment, and evaluation as a single system. - Is pragmatic: Optimizes for model quality and customer outcomes over publications or theory. - Communicates clearly: Can translate between customer needs and internal technical teams, and push back when needed. THE WORK - Act as the technical owner for enterprise customer VLM post-training engagements. - Translate customer requirements into concrete multimodal post-training specifications and workflows. - Design and execute visual data generation, filtering, and quality assessment processes, including image-text pair curation, annotation pipelines, and synthetic data generation for visual tasks. - Run supervised fine-tuning, preference alignment, and reinforcement learning workflows for vision-language models. - Design task-specific evaluations for visual understanding, grounding, OCR, document parsing, and other multimodal capabilities. Interpret results and feed learnings back into core post-training pipelines. DESIRED EXPERIENCE Must-have: - Hands-on experience with data generation and evaluation for VLM or multimodal post-training. - Experience training or fine-tuning vision-language models using SFT, preference alignment, and/or RL. - Strong intuition for visual data quality, annotation design, and multimodal evaluation. - Familiarity with vision encoders, image-text architectures, and how visual representations interact with language model backbones. Nice-to-have: - Experience with visual grounding, document understanding, OCR, or video understanding tasks. - Experience contributing to shared or general-purpose multimodal post-training infrastructure. - Prior exposure to customer-facing or applied ML delivery environments. - Familiarity with alignment or RL techniques beyond basic supervised fine-tuning in the multimodal setting. WHAT SUCCESS LOOKS LIKE (YEAR ONE) - Independently owns and delivers enterprise VLM post-training projects with minimal oversight. - Is trusted by customers as the technical owner, demonstrating strong judgment and delivery quality on multimodal workloads. - Has made durable contributions to Liquid's general-purpose multimodal post-training pipelines by feeding applied learnings back into baseline model development. WHAT WE OFFER - Real ML work: You will fine-tune vision-language models, generate multimodal data, and ship solutions, not configure API calls. Your work feeds directly back into our core model development. - Compensation: Competitive base salary with equity in a unicorn-stage company. - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents. - Financial: 401(k) matching up to 4% of base pay. - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year.

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY LFM2.5-Audio is Liquid's end-to-end multimodal speech and text language model. At 1.5B parameters, it handles speech-to-speech conversation, ASR, and TTS without requiring separate components, making it uniquely suited for real-time, on-device deployment. We're now bringing this model to enterprise customers. The core challenge: teaching audio models to understand user intents and translate them into structured tool calls. Think voice-driven function calling, where a spoken request triggers the right API, extracts the right parameters, and confirms back to the user in natural speech. This role sits at the intersection of frontier audio models and real-world deployment. You'll own the applied post-training work that adapts LFM2.5-Audio for customer use cases end-to-end, from data generation through delivery. Unlike most roles that force a trade-off between customer impact and foundational work, this one gives you both: deep ownership over how audio models are adapted, evaluated, and shipped, and a direct line into the evolution of Liquid's post-training and audio stacks. If you care about data quality, evaluation, and making models actually work in production, this is a chance to shape how applied audio AI is done at a foundation model company. WHAT WE’RE LOOKING FOR We need someone who: - Takes ownership: Owns customer post-training projects end-to-end for audio workloads, from requirements through delivery and evaluation. - Thinks end-to-end: Can reason across audio data pipelines, speech-text alignment, model adaptation, and evaluation as a connected system. - Is pragmatic: Optimizes for model quality and customer outcomes over publications or theory. - Thrives under constraints: On-device, low-latency, memory-limited audio systems excite you. You see constraints as design parameters, not blockers. THE WORK - Act as the technical owner for enterprise audio post-training engagements. - Translate customer requirements into concrete post-training specifications and workflows for LFM2.5-Audio and future audio models. - Design and build function calling capabilities for audio models: training models to map spoken user intents to structured tool calls (API invocations, parameter extraction, confirmation flows). - Design and execute data generation pipelines for speech-to-speech and text-to-text training, including synthetic dialogue, function calling examples, and intent-action pairs. - Run supervised fine-tuning, preference alignment, and reinforcement learning workflows on audio language models. - Design task-specific evaluations for audio function calling (intent recognition accuracy, parameter extraction, end-to-end task completion) and feed learnings back into core post-training pipelines. DESIRED EXPERIENCE Must-have: - Hands-on experience with post-training for language models (SFT, preference alignment, and/or RL). - Experience with data generation and evaluation pipelines for LLM or audio model training. - Strong intuition for data quality and evaluation design. - Familiarity with function calling, tool use, or structured output training for language models. Nice-to-have: - Experience with speech or audio language models (speech-to-speech, ASR, TTS, or multimodal audio-text systems). - Prior exposure to customer-facing or applied ML delivery environments. - Experience with alignment or RL techniques beyond basic supervised fine-tuning. - Familiarity with on-device or low-latency inference constraints. WHAT SUCCESS LOOKS LIKE (YEAR ONE) - Independently owns and delivers enterprise audio post-training projects with minimal oversight. - Has built and shipped function calling capabilities that reliably translate spoken user intents into tool calls for production use cases. - Is trusted by customers as the technical owner, demonstrating strong judgment and delivery quality. - Has made durable contributions to Liquid's general-purpose post-training and audio pipelines by feeding applied learnings back into baseline model development. WHAT WE OFFER - Real ML work: You will fine-tune audio and speech models, build audio data pipelines, and ship solutions to enterprise customers under real-time on-device constraints. - Compensation: Competitive base salary with equity in a unicorn-stage company - Health: We pay 100% of medical, dental, and vision premiums for employees and dependents - Financial: 401(k) matching up to 4% of base pay - Time Off: Unlimited PTO plus company-wide Refill Days throughout the year

ABOUT LIQUID AI Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there. THE OPPORTUNITY As an Applied ML Engineer on the Japan team, you will own the technical path from a customer's problem to a deployed AI solution. You will work directly with technical teams at leading companies in Japan and collaborate closely with Liquid's research, model, inference, and product teams in Japan and at our US headquarters. The center of gravity for this role is deployment: making models perform reliably in the environments where customers actually need them. Depending on the problem, that may also require post-training, careful evaluation, inference optimization, or changes elsewhere in the ML system. Your expertise will shine across these boundaries. You will join a growing Japan organization, with the autonomy to own important customer work and the backing of the teams building Liquid's core technology. This is an opportunity to shape how advanced, efficient foundation models are deployed in one of Liquid's fastest-growing markets. WHAT WE'RE LOOKING FOR We need someone who: - Owns outcomes end to end: You take responsibility from technical discovery through implementation, validation, optimization, and deployment. - Enjoys technical customer work: You can explore a problem with a customer's engineers, challenge assumptions constructively, and turn an ambiguous need into a sound technical plan. - Builds for the real environment: You treat latency, memory, compute, privacy, reliability, and maintainability as part of the ML problem. - Works with rigor: You move quickly while using strong baselines, profiling, careful evaluation, and disciplined error analysis to decide what works. - Collaborates across boundaries: You communicate clearly across customers, the Japan team, and globally distributed research and engineering teams. THE WORK - Own applied ML projects for customers in Japan, from technical discovery and scoping through production deployment. - Integrate, profile, and optimize model inference to meet concrete requirements for latency, throughput, memory, power, cost, and reliability. - Build the surrounding software needed to turn a model into a robust product capability, including data pipelines, evaluation systems, serving components, and reference implementations. - Fine-tune or post-train models when needed using techniques such as supervised fine-tuning, parameter-efficient fine-tuning, and preference optimization. - Design task-specific evaluations, conduct systematic error analysis, and iterate across data, models, inference, and system design. - Work directly with customer engineering teams during design, integration, testing, and rollout, including occasional on-site work. - Turn lessons from individual deployments into reusable tooling and feedback that improves Liquid's models, inference stack, documentation, and product roadmap. DESIRED EXPERIENCE Must-have: - Strong engineering skills and experience building, testing, and shipping production-quality ML systems. - Hands-on experience deploying modern language models, multimodal models, or other deep learning systems beyond a notebook or API proof of concept. - Experience with model serving, performance profiling, or inference optimization, and the judgment to balance model quality with system constraints. - Proficiency with the open-source ML ecosystem. - Experience designing evaluations, analyzing model failures, and using the results to drive measurable improvements. - Comfort leading technical discussions with customers and translating ambiguous requirements into shipped systems. - Professional proficiency in English, including the ability to collaborate on complex technical work with global teams. - Experience leveraging agents to amplify your own work. Nice-to-have: - Working proficiency in Japanese. - Experience with LLM post-training methods. - Experience with inference and deployment frameworks such as vLLM, SGLang, llama.cpp, ONNX Runtime, MLX. - Familiarity with quantization, hardware-aware optimization, or deployment on mobile, embedded, automotive, or other edge platforms. - Experience with multimodal systems involving text, vision, audio, or sensor data. - Experience delivering ML systems for enterprise or regulated environments. We care about demonstrated ability, learning speed, and engineering judgment. Degrees and publications are not required, and we encourage you to apply even if your experience does not match every item above. WHAT SUCCESS LOOKS LIKE (YEAR ONE) - You independently own important customer workstreams and earn trust through technical depth, clear communication, and reliable execution. - You move high-value use cases from ambiguous requirements to production-ready deployments under real performance constraints. - You deliver measurable improvements in model quality, latency, memory use, throughput, cost, or another outcome that matters to the customer. - You create reusable deployment, evaluation, or adaptation workflows that make subsequent projects faster and more reliable. - Your insights from the field materially influence Liquid's models, inference stack, tooling, or roadmap. HOW WE WORK This is a primarily remote role for candidates already based in Japan. We typically meet in our Tokyo office once or twice a week. Working hours are fully flexible, with an expectation that you can collaborate effectively with customers in Japan and colleagues in other time zones. The role includes travel to the United States and to relevant conferences, as well as occasional travel within Japan to work with customers in person. WHAT WE OFFER - High-impact work: Take foundation models from promising prototypes to systems that perform reliably in the real world. - Distinctive technical problems: Work on deployment challenges where efficiency, privacy, latency, and reliability are first-class requirements. - Direct access to core teams: Collaborate closely with the researchers and engineers building Liquid's models and inference technology. - Ownership in a growing market: Help define our technical approach in Japan at a time of strong customer demand and rapid growth. - Supportive & growth-driven culture: Thrive in a collaborative, feedback-rich environment that prioritizes continuous learning, mentorship, and personal career development. - Flexibility: Primarily remote work, flexible working hours, and unlimited paid time off. - Competitive rewards: Competitive salary, equity in a unicorn-stage company, and standard benefits for employees in Japan.