back to experience

Generative AI Engineer

Owned production LLM inference routing and evaluation for customer-support automation.

Organization
Concentrix + Webhelp
Period
Feb 2024 – Jul 2024
Location
Newark, CA

What I worked on

  • Owned the production LLM inference routing layer for 50K daily requests across 3 foundation models on AWS Bedrock and SageMaker; used LiteLLM, provider fallback, and latency-aware routing to cut p95 latency from 4s to 1.5s.
  • Built production LLM evaluation and safety gates with CloudWatch logging, hallucination checks, drift monitoring, and a 500-case test set; reduced incidents by 42% and MTTD by 35%.
  • Reduced monthly inference spend from $45K to $37K through cost-aware routing, prompt caching, and provider fallback while maintaining 95%+ task success.

Technical context

Python · AWS Bedrock · SageMaker · LiteLLM · CloudWatch · LLM evaluation

Read the LLM routing case study