Vengalarao Pachava

About

Backend/applied AI engineer from Cloud Rack Systems who built production GenAI/RAG and data platforms on Azure/Databricks at enterprise scale (2.5M records/day). Known for making LLM systems behave like deterministic services via strict retrieval contracts, citation-based validation, and strong observability—shipping a knowledge assistant used daily by 50+ users while driving hallucinations near zero and materially improving latency and cost.

Experience

Education

Key Strengths

• Built and operated Azure-based lakehouse + AI retrieval platform processing 2.5M records/day from 15+ sources
• Designed production RAG systems with strict retrieval contracts and deterministic execution behavior
• Implemented post-generation validation with citation/claim-to-evidence checks; grounded response rate improved to 96% and hallucinations near zero
• Shipped GenAI knowledge assistant used daily by 50+ users; reduced lookup time from 1–2 hours to <30 seconds
• Owned end-to-end GenAI feature delivery (embedding/indexing, retrieval logic, prompts, eval, monitoring, cost optimization)
• Improved LLM cost efficiency via semantic caching and token trimming
• Designed robust agent/workflow architecture with planning/execution/verification phases and explicit stop conditions (NO_EVIDENCE)
• Strong production engineering practices in Python (layered architecture, typing/Pydantic contracts, async IO, testing, CI, observability)
• PostgreSQL schema design for operational analytics + AI metadata with integrity constraints and migrations
• Database performance tuning: composite indexes, filter pushdown, date partitioning, and materialized views; reduced query latency ~5s to <800ms
• Designed guardrails for AI production deployments (drift/regression monitoring, audit logs, rollback gates, index integrity checks)