.png){kind=logo}
Harshal Sawant
Senior AI Engineer specializing in LLMs, RAG, and MLOps on multi-cloud
Connect with Harshal
Harshal already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
Built and productionized a secure internal RAG-based AI assistant (LangChain/FastAPI/FAISS on GCP), tackling real-world issues like latency, retrieval speed, and hallucinations—delivering 25% faster retrieval and 99.9% uptime. Also implemented scalable, reliable ML retraining orchestration with AWS Step Functions/SageMaker/Lambda and partners closely with compliance analysts to iteratively refine prompts and outputs to meet governance standards.
Experience
Key Strengths
- Built and deployed a secure internal AI assistant using production RAG architecture (LangChain/FastAPI/FAISS/GCP)
- Improved system performance with 25% faster retrieval and achieved 99.9% uptime
- Reduced hallucinations via better chunking and citation-based prompting
- Lowered latency under load by moving LLM serving to vLLM
- Designed reliable, scalable ML orchestration with AWS Step Functions (validation gates, retries, rollback; event-driven triggers; SageMaker autoscaling)
- Strong evaluation discipline for agents/workflows (unit tests, simulations, edge cases; metrics-driven debugging)
- Diagnosed and fixed stale-index issues in RAG by tracking retrieval hits/accuracy and repairing embedding refresh pipeline
- Effective collaboration with non-technical compliance stakeholders using diagrams, prototypes, and iterative feedback
Discover more candidates like Harshal
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent