Research Engineer, Post-Training Inference

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 ship AI products. We are looking for an engineer with strong experience in machine learning and solid foundations in math and computer science to join our growing Post-Training team at Baseten. THE ROLE Custom models are instrumental to the success of Baseten customers. By inference volume, the overwhelming majority of traffic at Baseten is to and from models that have been post-trained in some way, whether that be through reinforcement learning, supervised finetuning, a recent technique from the literature, or an in-house research technique from Baseten. The Post-Training team is responsible for the success of our customers’ post-trained models, and we employ a wide array of techniques to produce models that are more efficient and higher quality than even the biggest closed source models for the customer’s specific needs. Your role as a research engineer is to build the in-house tooling to support all of this. We care about training a wide spectrum of different model architectures with a variety of techniques efficiently and at scale. At times this involves zooming deep into a particular technical topic, but more often it involves working across the stack as a whole - systems-level concepts like Kubernetes, cgroups, storage systems, and networking topologies, as well as PyTorch distributed tensor computation, and GPU kernels. The Manifesto: https://labs.baseten.co/manifesto RECENT RESEARCH Dense, on-policy or both? Repeated kv cache for

About the Team The Post-Training Frontiers team is responsible for training the frontier agents OpenAI ships to the world (GPT-Next). We train the flagship agentic models behind Codex, ChatGPT, and the API through large-scale reinforcement learning. The team’s work spans four areas. First, execution and science: working with teams across OpenAI to decide what can go into the final model and how, using scientific experiments and evals that are representative of the final pipeline so issues can be recognized early. Second, RL scaling: executing the final large-scale reinforcement learning run, making sure GPUs are used efficiently and training stays healthy. Third, research: improving horizontal capabilities like instruction following, factuality, memory, and multi-agent behavior, where the team’s broad visibility helps identify cross-cutting improvements across teams and domains. Fourth, engineering: maintaining the infrastructure stack and internal tools to ensure that both the final run and all integrations go as smoothly as possible and that the systems are easy to work with. About the Role This role focuses on keeping our frontier RL training runs fast, reliable, and unblocked. You will work across engineering and infrastructure problems as they emerge, from scaling and orchestration issues to inference bottlenecks, numerical problems, and hardware failures, as well as supporting large horizontal integrations in the big run, like multi-agent capabilities or memory. This is a role for a strong generalist who quickly learns anything needed for the task, has high attention to detail, debugs deeply, and is motivated by fixing the highest-impact problem in front of the team. In this role, you will: Keep large-scale async RL training runs moving by jumping into the most urgent engineering and infrastructure problems. Debug issues across training systems, inference, orchestration, scaling, and distributed infrastructure. Improve the reliability and efficiency of RL trai

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 ship AI products. THE ROLE Forward Deployed Engineers work directly with the largest and fastest-growing AI companies in the world, owning their technical outcomes on Baseten and taking on the hardest problems in serving and improving models at scale. The work spans the model lifecycle: inference, post-training, and the systems that tighten the loop between them. Act as each account's de facto CTO on Baseten, with final accountability for how their workloads are designed, run, and scaled. Take customer objectives from vague to shipped: frame the problem, define the spec and success criteria, build the PoC, and carry it through to production quickly, using the right tools for the problem. Design the evals and benchmarks that isolate where quality or performance falls short, then close the gap yourself, whether that means optimizing inference, improving the model through post-training, or reworking the eval itself. Be the first responder to mission-critical failures including triage, owning the fix directly or route to the owning team and stay accountable until it ships. Build internal systems so that each engagement is faster than the last. This includes tooling and automation for eval and deployment infrastructure, and the recipes and reference implementations that make the product more self-serve. Shape the product itself, channeling what your accounts need into the roadmap and shipping fixes and features into Baseten's codebase yourself

About the Team Our team builds classification and evaluation systems that give post-training and research teams fast, reliable feedback on model behavior. We combine machine learning and production engineering to make these capabilities accurate, efficient, and useful at scale, helping researchers assess changes and improve models and products. We work closely with research teams to develop useful measurements and make them practical to run. Our work spans classifier development, rigorous evaluation, and efficient production inference, with ownership from initial experimentation through deployment and continued improvement. About the Role We’re looking for a machine learning engineer to develop new classification and evaluation capabilities and bring them into production. You’ll help turn open-ended questions about model behavior into well-defined ML problems, build methods that address them, and establish how reliably those methods work. This role combines traditional machine learning, model evaluation, and real-time engineering. You’ll own classifiers throughout their lifecycle, including supervised fine-tuning, distillation, evaluation, deployment, and production inference. You’ll work closely with researchers to identify valuable problems and with software engineers to make promising approaches useful at scale. In this role, you will: Build, deploy, and operate production classification and evaluation systems that provide useful feedback to post-training and research teams. Optimize production inference, including model serving, batching, caching, throughput, latency, compute utilization, and cost. Train and optimize classifiers using supervised fine-tuning and distillation, validating quality and efficiency before deployment. Own the reliability and observability of deployed classifiers, diagnose production issues, and improve their performance over time. Design rigorous evaluations that assess accuracy, calibration, generalization, and potential measurement bi

About the Team Our team builds classification and evaluation systems that give post-training and research teams fast, reliable feedback on model behavior. We combine machine learning and production engineering to make these capabilities accurate, efficient, and useful at scale, helping researchers assess changes and improve models and products. We work closely with researchers and data scientists and have broad ownership of the software that makes these capabilities available. Our work spans real-time systems, data engineering, model serving, and tools for experimentation, with careful attention to performance, reliability, and privacy. About the Role We’re looking for a software engineer to build and evolve our classification and evaluation platform. You’ll develop services, online and offline pipelines, and tools that help researchers and data scientists turn an idea into a working measurement and iterate on it quickly. This role combines software and data engineering with hands-on production ownership. You’ll shape the architecture, deliver new capabilities, improve inference efficiency, and keep the system dependable as workloads and requirements evolve. Strong candidates will bring deep software engineering expertise and an interest in building a technical product in close partnership with its users. In this role, you will: Design and build software services and APIs for configuring, running, and managing classifiers and evaluations. Build and operate online and offline classification pipelines that support reliable, efficient execution at scale. Improve model-serving performance, including throughput, latency, concurrency, and compute utilization. Build tools that make experimentation, debugging, and iteration straightforward for researchers and data scientists. Own production reliability, observability, and incident response, and reduce the operational burden of running the platform. Partner with machine learning engineers to bring new classification capabili

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 ship AI products. THE ROLE Are you the person on your team who builds the agent everyone else ends up using? We're looking for an AI Engineer to join our Training Product team and do that at Baseten. You'll build AI-driven product features for the customers training and post-training frontier models on our platform, and you'll raise the ceiling on how Baseten itself uses AI internally, turning manual workflows into agentic ones that make every other team faster. You'll work directly with our research engineers to scope and build products, taking ideas from a research loop that already works internally to something customers can run themselves. This is a hands-on role with real autonomy. You'll pick the problems worth solving, build the harnesses, execution flows, and guardrails that make AI systems reliable, and own the results. If you've been shipping agents and want that to be the job, let's talk. EXAMPLE INITIATIVES Take a look at these blog posts written by members of our team: Baseten Training: an autoresearch substrate Introducing Baseten Loops Harnesses are everything. Here's how to optimize yours. Building with NVIDIA Nemotron 3 Ultra and LangChain Deep Agents Code on Baseten RESPONSIBILITIES Build and ship agentic product experiences, including chat-style and assistant-like interfaces, from prototype to GA. Design the harnesses, execution flows, and guardrails that make AI systems reliable in production. Build internal automation