Forward Deployed Engineer (Training)
Core
Act as a de facto CTO for AI companies, owning their technical outcomes on Baseten for model serving, inference, and post-training at scale.
Role type
Senior IC Forward Deployed Engineer (AI Infrastructure)
Builds
Production-grade AI inference and post-training systems for enterprise customers
Domain
AI Infrastructure / Large Language Model Serving
Deliverable
production ML models
Required skills
Debugging complex production issues, owning ambiguous technical problems, clear communication on complex technical topics, operational depth in incident response, sequencing work across multiple accounts
Preferred skills
Depth in core infrastructure domains (storage, networking), operating distributed compute platforms (Kubernetes, Slurm, Ray), understanding LLM architectures and inference engines (vLLM, TensorRT-LLM, SGLang), profiling and optimizing GPU workloads, hands-on experience with post-training techniques (SFT, RL), fluency in tensor computation libraries (PyTorch, JAX)
Technologies
Kubernetes, Slurm, Ray, vLLM, TensorRT-LLM, SGLang, PyTorch, JAX, InfiniBand, RoCE
Responsibilities
Design and scale customer workloads, take customer objectives from vague to shipped via PoC and production, design evals and benchmarks to close quality gaps, respond to mission-critical failures, build internal tooling and automation, shape the product roadmap based on customer needs
Seniority
Senior, hands-on IC
