Senior AI Engineer · Agentic Systems & AI Platforms · Technical Lead
Building production-grade AI systems that work in real-world enterprise environments.
I’m a Technical Lead and Senior AI Engineer with 20+ years of experience designing, building, and operating large-scale enterprise systems, now focused on agentic AI, LLM platforms, and Retrieval-Augmented Generation (RAG).
My work lives at the intersection of AI innovation and engineering rigor—moving systems from prototypes to secure, observable, and reliable production deployments. I enjoy solving complex, high-stakes problems where AI must integrate cleanly with existing platforms, APIs, and operational constraints.
I’m known for IC+ leadership: owning architecture end-to-end, mentoring engineers, and delivering systems that create measurable business value.
- Agentic and tool-using AI systems
- LLM orchestration, evaluation, and guardrails
- Enterprise RAG with strong grounding and trust guarantees
- Cloud-native, event-driven AI platforms
- Production reliability, observability, and security
Agentic GitHub repository exploration and analysis
Repo Navigator AI is a tool-driven, multi-agent system that helps engineers explore and understand GitHub repositories by interacting directly with the GitHub API, rather than pre-ingesting or indexing code.
What it does
- Uses specialized agents to navigate repositories, files, commits, and pull requests
- Dynamically retrieves code and metadata via GitHub API tools
- Provides source-aware, explainable responses grounded in live repository state
- Designed for architectural reasoning, onboarding, and codebase comprehension
Why it matters
- Avoids stale indexes and heavy ingestion pipelines
- Mirrors how engineers actually explore real repositories
- Emphasizes correctness, traceability, and tool-based reasoning
👉 Repo: https://github.com/VandanaJn/Repo-Navigator-AI
Source-grounded RAG assistant with enterprise guardrails
Sage AI is a Retrieval-Augmented Generation (RAG) system designed to deliver trustworthy, source-backed answers over curated knowledge bases.
Highlights
- Emphasis on grounding, citation, and hallucination reduction
- Designed with enterprise-grade guardrails and secure data handling
- Focus on observability, prompt discipline, and predictable behavior
- Built to be extended into production workflows rather than demos
Sage AI reflects my approach to RAG: accuracy over cleverness, and systems that can be safely deployed in real organizations.
AI & Agentic Systems
Multi-agent architectures, tool calling, RAG, LLM orchestration and evaluation
Google Vertex AI (ADK), AWS Bedrock, LangChain, LlamaIndex
Cloud & Distributed Platforms
AWS (Lambda, SNS/SQS, API Gateway, CDK, S3, DynamoDB)
GCP (Vertex AI, Cloud Run)
Event-driven and serverless architectures
Backend & Data
Python, FastAPI, C#, SQL, PostgreSQL, Oracle, DynamoDB
High-throughput, low-latency enterprise systems
Platform Engineering
Docker, CI/CD (GitHub Actions, CodeBuild, CodePipeline)
Infrastructure as Code, observability, secure logging, authN/authZ
- Production-first AI: reliability, security, and observability are mandatory
- Tooling over guessing: agents should act, not hallucinate
- Systems thinking: AI is part of a platform, not a feature
- Ownership mindset: design it, ship it, support it
If you’re building enterprise AI platforms, agentic systems, or customer-facing AI deployments, I’m always happy to exchange ideas.