About Us We are a small team of researchers and engineers focused on pushing the frontier of what agents are capable of. We believe the next leap in agent performance will come not from bigger general models, but from specialization and continual learning: agents that get better at their specific job the more they do it. We build the systems that make this possible, turning real-world usage into lasting improvement.
Summary We are looking for an infrastructure lead to own the systems that train and serve our models. You will be responsible for our reinforcement learning stack, which runs rollouts and trains models across large GPU clusters for days at a time, and for the inference platform that serves many specialized models to customers. You will set the technical direction for both, build them hands on in the first months, and then hire and lead the team that runs them. You will work closely with our research leads and report to our VP of Engineering.
Responsibilities
Own the end to end RL training system: rollout generation, reward and verifier services, policy updates, and evaluation.
Own production inference: serving many fine tuned models on shared hardware, safe rollouts of new versions, and deployments in customer clouds and VPCs.
Keep long training runs healthy without manual intervention, and make researchers faster.
Improve throughput and GPU utilization across training and inference, and plan compute with the research team.
Integrate inference engines such as vLLM or SGLang into the training loop and optimize them for RL workloads.
Make the hard calls on architecture, vendors, and clouds, and explain them clearly to the team.
Hire, grow, and lead the infrastructure team as we scale.
Set engineering standards, on call practices, and the quality bar for the group.
Qualifications
7+ years building ML or distributed systems infrastructure, including 2+ years leading an engineering team.
A track record of leading teams that built and ran large scale ML training or inference systems.
Deep experience with distributed systems and training large models across many GPUs.
Hands on experience with PyTorch and distributed training frameworks such as FSDP, DeepSpeed, Megatron-LM, or Ray.
Production experience with LLM serving, including inference engines such as vLLM, SGLang, or TensorRT-LLM.
Solid understanding of RL workflows for language models and agents.
Strong Python skills. C++ or CUDA is a plus.
Experience hiring and growing strong engineers.
Comfortable with ambiguity and able to own large problems end to end.
Nice to Have
Led RL, training, or inference infrastructure at a frontier lab or AI infrastructure company.
Contributions to open source RL, training, or inference frameworks.
Experience with Kubernetes, Slurm, or other cluster schedulers at scale.
Experience with agent environments, code execution sandboxes, or continual learning systems.
Job details are sourced from the employer's original posting.
Open job postingAbout the company
Noeri develops AI-powered solutions for enterprise applications.