Dhanalakshmi Jammisetti

About

LLM/RAG engineer who has shipped production systems in high-stakes domains (fraud analytics at Mastercard and security compliance as a CI/CD gate). Strong focus on reliability: hybrid retrieval for latency, citation-backed outputs for trust, and code-driven eval/regression pipelines using golden datasets. Also built scalable OCR-based ingestion for messy classroom artifacts (handwriting, PDFs, whiteboard photos) using Go/Python and cloud services.

Experience

Education

Key Strengths

• Owned end-to-end RAG feature for fraud analytics (static reporting → conversational insights)
• Performance/quality tradeoff execution (implemented hybrid search to improve retrieval latency)
• Built trust mechanisms for high-stakes LLM apps (citations mapped to source rows/metadata IDs in Postgres)
• Shipped production compliance auditor integrated as a CI/CD gate using RAG over enterprise docs
• Code-driven LLM evaluation and regression prevention (golden dataset + test assertions run on prompt/retrieval changes)
• Designed multi-metric eval loops (faithfulness, coherence, actionability) with human review feedback incorporation
• Built scalable ingestion pipeline for messy real-world artifacts (handwriting, PDFs, photos) with OCR + normalization
• Scaled pipelines with batching/autoscaling and cost controls (optimized OCR settings, spot instances)
• Owned end-to-end delivery of a real-time inventory dashboard (frontend, backend, deployment, monitoring)
• Real-time systems implementation (live updates/latency monitoring) using SignalR and microservices
• Cloud-native deployment automation with containers and Kubernetes (Docker + AKS) and CI/CD pipelines
• Designed secure APIs with OAuth 2.0 and RBAC integrated with Azure AD
• Built modern full-stack web apps (React/TypeScript + Node/Express + Postgres) with layered architecture (controllers/services/repository)
• Production AWS operations experience across ECS, Lambda, API Gateway, ALB, RDS, DynamoDB, S3, CloudFront with multi-environment setup
• High-ambiguity execution: shipped MVP workflows quickly via stakeholder demos and iterative releases
• Maintainability improvements via modular UI component library, backend service-layer refactor, and enforced code standards (ESLint/Prettier)
• Reliable data ingestion/ETL patterns: idempotency, retries with exponential backoff, and CloudWatch alerting