The role
- COMP
- $176K - $224K
- EQUITY
- Competitive equity
- LOCATION
- New York · San Francisco · South Bay Area
- WORKPLACE
- Remote
- EXPERIENCE
- 3+ years
- VISA
- None, Visa transfers
- STACK
- Python, Kubernetes, AWS, Azure, GCP
- INDUSTRY
- AI, Software Development
/roles — ROLE_482
AI Field Engineer - Enterprise
Late-stage AI inference platform helping large enterprises run and fine-tune open-source models in production
The company
A late-stage AI infrastructure company whose cloud platform lets teams run, fine-tune, and scale open-source models in production, used by many well-known consumer and software companies.
- STAGE
- public-scale
- FUNDING
- public-scale
- TEAM
- 200+ people
- BACKING
- backed by Sequoia, Benchmark, Lightspeed
JD — the work
About the role
A field engineering role for someone who wants to sit inside large enterprise accounts and turn difficult generative AI problems into systems that run in production, quickly. You would combine real engineering depth with the presence to win the confidence of senior leaders across big organizations, carrying deals from the first discovery conversation to a live deployment. The work is heavily customer-facing, remote-friendly, and what you learn flows straight to the product and engineering teams.
What you'll do
- Lead discovery sessions, define proof-of-concept scope, and benchmark load and quality to pick the right model setup
- Build prototypes and production integrations yourself inside customer environments, working within their infrastructure and security rules
- Advise on picking models, evaluation design, and fine-tuning with SFT, DPO, or RFT as customers scale open models
- Handle multi-party enterprise relationships, spot internal champions, and work through organizational dynamics to keep deals on track
- Report repeated customer problems and common deployment patterns to engineering so they shape the roadmap
What they're looking for
- 3+ years in customer-facing or pre-sales field engineering (forward-deployed, applied AI, solutions architecture); purely internal engineering is not enough
- Deep practical experience training or serving LLMs, including open-model stacks like vLLM, SGLang, or TensorRT-LLM
- Hands-on fine-tuning with SFT at the least; experience limited to closed models or API wrappers will not qualify
- Proof that you have deployed POCs or MVPs into another company's production systems, beyond advisory work
- Solid Python plus GPU and cloud infrastructure skills (AWS, GCP, or Azure), and working Kubernetes knowledge
- Executive presence: equally comfortable in a deep technical session with ML engineers and an architecture review with a VP
Nice to have
- Experience with DPO or RFT beyond standard supervised fine-tuning