Inference Engineer in New York

Who are we? Cohere is the leading security-first enterprise AI company. We build cutting-edge foundation AI models and end-to-end products that are designed to solve real-world business problems. We’re training and deploying frontier models for enterprises who are building AI systems. We believe that our work is instrumental to the widespread adoption of AI and we are looking for folks that want to be part of that. We obsess over what we build. Each one of us is responsible for contributing to increasing the capabilities of our models and the value they drive for our customers. Cohere is a team of researchers, engineers, designers, and more, who are all passionate about their craft. We are a global technology company headquartered in Toronto with key offices in London, New York City, San Francisco, Montreal, Paris, Berlin and Seoul. Join us! Why this role? Our team is a fast-growing group of committed researchers and engineers. The mission of the team is to build reliable machine learning systems and optimize audio inference serving efficiency using innovative techniques. As an engineer on this team, you will work on advancing core audio model serving metrics, including latency, throughput, and quality by diving deep into our systems, identifying bottlenecks, and delivering creative solutions for audio processing and streaming workloads. You’ll collaborate closely with both the training and serving infrastructure teams to ensure seamless integration between model development and deployment, with a special focus on real-time and streaming audio inference. Please Note: We have offices in Toronto, Montreal, San Francisco, New York, Paris, Seoul and London. We embrace a remote-friendly environment, and as part of this approach, we strategically distribute teams based on interests, expertise, and time zones to promote collaboration and flexibility. You'll find the Model Efficiency team concentrated in the EST and PST time zones, these are our preferred locations. You may

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role: Most of the value of owning a model shows up at serving time. We're building a platform that covers the whole life of an LLM -- train it, deploy it, observe it -- and inference is where teams feel the difference every day. We already run elastic inference, sandboxes, distributed volumes, and multi-node training, and we control the infrastructure underneath, so the serving stack is ours to shape rather than something we resell. You will do hands-on inference research at Modal, working with the research lead to pick high-impact bets and owning them end to end. The bets that matter most are the ones that move cost per token and tail latency on the workloads our customers actually run. What you'll do: Own end-to-end inference research bets: speculative decoding, disaggregated prefill/decode, quantization (FP8, INT4), KV-cache and memory management, autoscaling for spik

ABOUT BASETEN Baseten powers mission-critical inference for the world's most dynamic AI companies, like Cursor, Notion, OpenEvidence, Abridge, Clay, Gamma and Writer. By uniting applied AI research, flexible infrastructure, and seamless developer tooling, we enable companies operating at the frontier of AI to bring cutting-edge models into production. We're growing quickly and recently raised our $1.5B Series F , led by Altimeter Capital, Conviction Partners, and Spark Capital. Join us and help build the platform engineers turn to to ship AI products. THE ROLE Baseten's GTM org is in hyper growth. As it grows and matures, the needs of the GTM stack get more sophisticated with scale — and this team exists to stay ahead of those needs. Our GTM Engineering team plays a critical role in building and maintaining the connective tissue across Baseten’s GTM tools, processes, and user experience for the field. The GTM tooling landscape is changing fast, and the teams that win are the ones that adapt and iterate the fastest. This role exists to make sure Baseten is one of them. You'll design, build, and ship AI-powered workflows that scale our GTM functions as a competitive advantage. We want someone who can walk in, audit what we have, identify what we're missing, and start shipping fast. You know when to reach for Clay and when to build something custom in Claude Code. You think two to three steps ahead about how the thing you build today fits into the broader systems architecture tomorrow. And you bring a point of view — on our stack, on what we should be building, and on where AI can do something low-code tooling simply can't. RESPONSIBILITIES Ship AI-powered workflows for the field — build the agents and automations that give reps and managers real leverage, off-loading the manual and repetitive work. Reach for AI where it does something low-code can't. Get insights in front of reps — turn Salesforce, warehouse, and usage data into the dashboards, scores, and alerts reps

Who are we? Cohere is the leading security-first enterprise AI company. We build cutting-edge foundation AI models and end-to-end products that are designed to solve real-world business problems. We’re training and deploying frontier models for enterprises who are building AI systems. We believe that our work is instrumental to the widespread adoption of AI and we are looking for folks that want to be part of that. We obsess over what we build. Each one of us is responsible for contributing to increasing the capabilities of our models and the value they drive for our customers. Cohere is a team of researchers, engineers, designers, and more, who are all passionate about their craft. We are a global technology company headquartered in Toronto with key offices in London, New York City, San Francisco, Montreal, Paris, Berlin and Seoul. Join us! Why this role? Large Language Models (LLMs) continue to push the boundaries of what AI systems can do — but inference is still the bottleneck. The Model Efficiency team is responsible for pushing the limits of LLM inference efficiency across our foundation models. We explore and ship breakthroughs across the model execution stack, including: model architecture and MoE routing optimization decoding and inference-time algorithm improvements software/hardware co-design for GPU acceleration performance optimization without compromising model quality Please Note: We have offices in Toronto, Montreal, San Francisco, New York, Paris, Seoul and London. We embrace a remote-friendly environment, and as part of this approach, we strategically distribute teams based on interests, expertise, and time zones to promote collaboration and flexibility. You'll find the Model Efficiency team concentrated in the EST and PST time zones, these are our preferred locations. As a Staff Research Engineer, you will develop, prototype, and deploy techniques that materially improve how fast and efficiently our models run in production. You may be a good fit

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. About Modal Data: We’re growing our Data team and are looking for our first few key hires to build self-serve data tools and drive business strategy in the right direction. The mission of the Modal Data team is to make it easy to track company goals, make evidence-backed decisions, and prioritize the right work. We do this via: Self-serve AI analytics tools (Hex, Snowflake) Embedding with teams as a “data adviser”, providing strategic analysis and consulting What You'll Do: Contribute to building the most modern analytics stack in Data today to support AI-driven self-serve analysis, key metrics tracking, and external customer reporting Influence work on new products like LLM Inference Endpoints through product analytics tracking Identify millions of dollars of cost savings and optimization across our tools and financial operations Write data pipelines that power the operatio

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role: Modal considers high-quality documentation to be essential for developer experience, and we see docs becoming even more important as agents increasingly deploy and operate Modal Apps. We are looking for a content-minded engineer who will partner with our product teams to curate Modal’s technical documentation and maintain a high quality bar across multiple dimensions. Responsibilities: Thinking holistically about content architecture and how the docs should evolve as Modal introduces new products and features Innovating on novel documentation formats and delivery channels to optimize agent productivity, in collaboration with our Agent DX research team Developing content standards, style guides, and automated enforcement mechanisms to ensure consistent style and high quality Building and maintaining automated pipelines that will enforce the correctness of code examp

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role: We're looking for engineers with deep AI/ML and low-level systems experience who want to build the best technical support experience in the world. This isn't a traditional support role — it's an engineering role where you happen to be closest to our customers. You'll split your time roughly 50/50 between working directly with customers and shipping fixes, features, and automation that improve Modal for everyone. When you help a customer debug a training run, you'll also fix the underlying issue in the platform. When you notice ten customers hitting the same friction point, you'll build the tooling or automation that eliminates it entirely. This role is for people who solve problems, not people who answer tickets. The problems you encounter are deeply technical and arise from running some of the most demanding AI workloads in the world. You'll be a member of our eng

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role We're looking for an Engineering Manager to lead a team of highly experienced engineers building the infrastructure that powers Modal's serverless GPU platform. This is a hands-on leadership role — expect to split your time between technical contribution and people management depending on what the team needs. You'll set direction, remove blockers, and build a strong engineering culture as your team tackles hard problems in distributed computing, large-scale data handling, and performance optimization. Who You Are You're an experienced engineering leader who stays close to the work and builds alongside your team when it counts. You earn trust through technical depth, not title. You communicate clearly, help strong engineers move fast without cutting corners, and stay calm and pragmatic under pressure. You care as much about how your team gets to an answer as the answ

Salary range - $250k - $350k | Equity - up to 0.5% | In-person NYC About Datalab Datalab trains models that read documents reliably at scale. The world's most important information is trapped in PDFs, scans, and files that can't easily be parsed, and getting it out correctly matters. From frontier AI labs processing training data to Fortune 500s like Siemens extracting decades of engineering records, Datalab is where businesses turn to when extraction has to be right. We’re at an 8-figure run rate with a team of 7. Anthropic is a customer. And we have hundreds more across FAANG, frontier AI labs, healthcare, finance, government, and legal. Our tools, Chandra, Surya, Marker, and Lift, have 70,000+ GitHub stars and broad developer mindshare. We're backed by founding members of OpenAI, FAIR, and Hugging Face. Role Overview We're looking for a Research Engineer to own problems end to end across our models, inference service, and product. You won't just train a model and hand it off. You'll take it from training through benchmarking, into our inference stack, and work with the team to integrate it into our products. We're a small team that has shipped the current state of the art OCR model, Chandra. Our models collectively have 70k+ Github stars. Our tools are used internally at frontier AI labs like Anthropic, and Fortune 500 enterprises like Siemens. Our team focuses on training small, efficient models that outperform much larger LLMs on domain-specific tasks (like OCR, structured extraction, tables). We move fast, prioritize practical results, and build tools that are open, reproducible, and built to last. You'll test hypotheses quickly, iterate on results, and balance experimental rigor with shipping to customers. Day to day: A typical project might look like: identify a gap in extraction quality on long documents, train and benchmark a new model, optimize it for inference, and work with the team to ship it to users. Concretely: Train and evaluate models: Train task-

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role: We're looking for Forward Deployed ML Engineers who want to work at the intersection of deep technical work and direct customer impact. As an ML FDE, you'll partner with leading AI companies and foundation model labs to help them achieve state-of-the-art performance on their most demanding workloads — LLM serving, model training (SFT, RLHF), audio pipelines, scientific computing, and more. You're helping teams reach outcomes most engineers can't on their own. The FDE team today includes world-class software engineers, computational scientists, ML engineers, and former founders. We're looking for people with strong engineering fundamentals, deep curiosity across the AI stack, and energy for working directly with customers on hard problems. You will: Work hands-on with companies like Suno, Lovable, Cognition, and Meta to architect and optimize production AI workloads

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role: We're looking for Forward Deployed Engineers on our engineering team who want to work at the intersection of deep infrastructure work and direct customer impact. As an FDE, you'll partner with leading AI companies and foundation labs on cloud architecture, networking, storage, containerization, sandboxing, and more — helping them design and ship production infrastructure on Modal's platform. The FDE team today includes world-class software engineers, computational scientists, ML engineers, and former founders. We're looking for people with strong engineering fundamentals, deep curiosity across the infrastructure stack, and energy for working directly with customers on hard problems. You will: Work hands-on with companies like Suno, Lovable, Cognition, and Meta to architect and deploy massive-scale production workloads on Modal Lead technical discovery and architect

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role: Modal builds AI infrastructure products that developers love. That's how we grew so quickly and why word-of-mouth remains one of our most important channels today. From powering one of the largest vibe-coding platforms at Lovable to enabling teams like Ramp to build their own internal coding agents , Modal Sandboxes are used by developers to safely execute AI-generated code at scale. We're now hiring our first developer relations engineer focused on Modal Sandboxes. Whether it’s banger tweets , in-depth technical resources or long-form talks , we want to meet developers by any medium necessary and empower them to build and ship novel AI products. In this role, you will primarily be creating and distributing technical content that is unique, educational, and practical. This content will be the first Modal touchpoint for many of our users. We want to not only showcas

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role: We’re looking for an Infrastructure Security Engineer to design and secure the core systems that power our platform. This role focuses on building security directly into our infrastructure—from container isolation and orchestration to identity and secrets management in a multi-tenant, cloud-native environment. You’ll work closely with engineering teams to define secure primitives and ensure our platform is resilient, scalable, and trustworthy by design. This is a hands-on, deeply technical role focused on real systems, not compliance or policy. What You'll Do: Platform & Runtime Security Design and improve isolation mechanisms for multi-tenant workloads (containers, sandboxing, execution environments) Strengthen boundaries between customers, workloads, and internal systems Identify and mitigate risks in distributed, dynamic compute environments Container &

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role: At Modal, we sell cloud services atop which our customers run their critical production systems. As a rapidly growing new cloud infrastructure company, we seek to improve our reliability dramatically while scaling the size of our platform, customer base, and our team. This role is for people who are deep systems thinkers, love stacking nines, and thrive from making others move faster at scale. Responsibilities include: Identifying architectural changes to improve reliability and performance. Fostering a culture of reliability across Modal’s engineering organization. Defining and implementing operational processes such as deployments, upgrades, etc. Operating systems like Kubernetes, Postgres, Redis, etc. Participating in on-call rotations, and responding to production incidents. Requirements: 5+ years of experience writing high-quality production code. 2+ years of

About Us: AI needs a new infrastructure layer. We're building it at Modal. Every era of computing brought new workloads that previous infrastructure couldn't support: mainframes, databases, and the cloud. Each time, the company that rebuilt the layer underneath defined the decade. AI is no different, except it touches everything instead of one slice, and the window to build the layer underneath it is open right now. Our customers include category-defining companies like Lovable , Ramp , Cognition, DoorDash, and Suno. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so it's simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale. We recently raised a $355M Series C at a $4.65B valuation, led by General Catalyst and Redpoint Ventures. We've crossed $300M+ ARR and grown fivefold since September. Our team includes creators of popular open-source projects (e.g., Seaborn , Luig i ), academic researchers, international olympiad medalists, and experienced engineering and product leaders with decades of experience. The Role: We're looking for a Detection & Response Engineer to build the systems that help us identify, investigate, and respond to threats across our platform. This is an engineering role focused on automation. You'll build detections, investigation tooling, and response capabilities that scale with our infrastructure, using AI where it meaningfully improves signal, investigation speed, and operational effectiveness. You'll work closely with infrastructure, platform, and security engineers to ensure every incident makes the platform more resilient. What You'll Work On: Detection Engineering Design and build high-fidelity detections for attacks, abuse, and anomalous behavior across our infrastructure and production systems Continuously improve detections based on telemetry, threat intelligence, and lessons learned from incidents Improve visibility across cloud infrastruc