Research Engineer, Forge
Core
Build and improve post-training and evaluation workflows (CPT/SFT/RL/distillation), turning prototypes into repeatable production pipelines for client AI systems.
Role type
Senior IC research engineer (LLM training & infrastructure)
Builds
Reliable training and deployment workflows (Forge recipes), synthetic data pipelines, and large-scale ML systems for enterprise clients.
Domain
Artificial Intelligence / Large Language Models / Distributed Systems
Deliverable
production ML models
Required skills
Python engineering, PyTorch/JAX, distributed training, LLM post-training (fine-tuning, RL, distillation), system debugging, data pipeline development
Preferred skills
FSDP/DeepSpeed/Megatron, cluster orchestration (Kubernetes/SLURM), multi-billion parameter model training, open-source contributions
Technologies
PyTorch, JAX, Kubernetes, Ray, SLURM, DeepSpeed, FSDP
Responsibilities
Develop tools for synthetic data generation and data curation; Debug and harden large-scale ML systems (distributed training, checkpointing, observability); Improve Forge codebase via APIs and documentation; Partner with researchers to translate bottlenecks into system improvements; Ensure seamless deployment across diverse client environments (cloud/on-prem).
Seniority
Senior, hands-on IC