Lila Sciences
Associate Director/Director, Inference Engineering
$204K–$348K
Cambridge, MA USA +3 · Director · Full-time · Posted Oct 26, 2026
- Location
- Cambridge, MA USA
San Francisco, CA USA
Alewife, Cambridge, MA
San Francisco, CA
About the job
Leads the team and technical strategy for Lila’s model-serving platform, taking inference services from research and ML into reliable production at scale.
What you'll do
- Own inference services end to end, including API contracts, latency SLAs, throughput guarantees, and rate limits.
- Set serving architecture, reliability practices, and systems for model onboarding, safe deployment, evaluation, and observability.
- Lead and grow the engineering team, own the serving roadmap, and partner with ML, research, and product engineering.
What they ask for
- Deep understanding of LLM inference stacks, serving frameworks, batching, memory constraints, and latency/throughput tradeoffs.
- Experience owning a customer- or product-facing API service end to end, including SLAs, incident response, and reliability.
- Track record hiring, mentoring, and retaining senior technical talent and scaling engineering organizations.
- Ability to set technical strategy, exercise strong technical judgment, and align stakeholders across competing priorities.
Nice to have
- Hands-on experience with speculative decoding, quantization, or model parallelism strategies.
- Experience with serving evaluation, GPU capacity planning, or work in AI/ML startups or research-heavy organizations.
Pay
$204,000 — $348,000 USD (base salary)
Tools
vLLM, TensorRT-LLM, TGI, AWQ, GPTQ, FP8
AI focus
Leads production LLM inference and model serving, including architecture and deployment of models at scale.
Model-written summary · the original posting is authoritative