Anand Buruganahalli Rajanna

AI Engineer

Multi-Agent Systems · LLM Orchestration · RAG · Seattle, WA

↓ AI-powered search coming soon — for now, scroll for the answer

about

I build AI systems that work in production, not just in demos. At Amazon, I architected a multi-agent orchestration platform on Amazon Bedrock, now used by 6+ engineering teams. Outside of work, I build agents to go deeper — JobPilot, an autonomous job-application system, taught me more about guardrails, evaluation loops, and human-in-the-loop design than any course could. I care about systems that are reliable, not just impressive in a demo.

experience

Software Development Engineer — Amazon

Seattle, WA · Aug 2022 — Present

Architected a production multi-agent AI system on Amazon Bedrock — a central orchestration agent routing natural language queries across specialized sub-agents. Adopted by 6+ engineering teams, cutting mean debugging time by 30+ min per incident.

Owned large-scale logistics systems across 300+ fulfillment centers, including a serverless failure-analysis pipeline that improved time-to-diagnosis by 70%.

Senior Software Engineer — Capgemini India

Bengaluru, India · Aug 2018 — May 2020

Led a cross-functional team of 10 engineers delivering enterprise solutions; built automation reducing report generation time by 50%.

projects

JobPilot

view source →

Python · Claude API · SQLite · Telegram Bot API · Browser Automation · Ollama

An autonomous agent that monitors job postings, scores fit via LLM reasoning, tailors resumes per role, and executes browser-based applications with human-in-the-loop approval gates at every step that matters. Thin harness, fat skills — deterministic orchestration logic stays in Python, judgment calls go to the LLM.

agentic guardrails human-in-the-loop

job-match-ai

view source →

Python · ChromaDB · Sentence-Transformers · FastAPI

An end-to-end RAG pipeline embedding job descriptions and candidate profiles, storing in ChromaDB, retrieving best-fit matches via semantic similarity, exposed as a REST API with basic drift monitoring.

RAG embeddings vector-search

contact

Email LinkedIn GitHub