Skip to content
View chiranjeevigundu's full-sized avatar

Highlights

  • Pro

Block or report chiranjeevigundu

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
chiranjeevigundu/README.md

Hi 👋, I'm Chiranjeevi Gundu

AI Engineer | GenAI & Multi-Agent Systems | RAG & LLM Apps

AI Engineer and systems architect with an M.S. in Computer Science (AI track) and 4 years of experience taking AI from prototype to production.

Portfolio LinkedIn Email


🚀 About Me

I design end-to-end GenAI systems — multi-agent orchestration, RAG pipelines, and model-routing layers — and ship them with backend rigor: typed APIs, sound data modeling, automated tests, CI/CD, and secure cloud and on-prem deployment on AWS and Azure.

  • 🔭 Currently: AI Engineer @ Axitem Software Solution Inc. — RAG document-search chatbots and LangChain/LangGraph agent workflows for enterprise clients
  • 🌱 Focused on multi-agent orchestration (LangGraph, CrewAI), RAG grounding with citations, and self-hosted LLMs (Llama 3, Mistral via Ollama)
  • 💬 Ask me about shipping LLM apps to production, vector search (Pinecone, pgvector), or human-in-the-loop agent safety
  • 📈 Proven impact: 40% faster ticket triage · 60% faster queries · 50% higher user adoption
  • 📫 Reach me: chiranjeevigundu1@gmail.com

🔥 Featured Project

Self-hosted, multi-user AI platform built end to end — data model, agent layer, ingestion, REST API, and iPhone app — running as seven Docker Compose microservices on FastAPI and Postgres 16 with pgvector (MIT licensed).

  • Multi-agent organization: a CEO agent decomposes goals and delegates to four specialist agent teams, with explicit step limits and a full audit trail
  • Human-in-the-loop safety: read-only analysis runs autonomously; any real-world side effect (money, email) becomes a proposal the user must approve
  • Tiered model routing over LiteLLM (OpenRouter, GPT-4o-mini, local Llama 3 via Ollama) — cost, capability, and privacy each map to the right model with zero code changes
  • Multi-tenant isolation with Postgres Row-Level Security, JWT auth, and Fernet-encrypted credentials
  • Shipped solo with 149 automated tests, GitHub Actions CI, and ingestion pipelines that turn emails, PDFs, and bank transactions into reviewable to-dos

🛠️ Technical Skills

Languages & ML — Python · SQL · JavaScript · TypeScript · Bash · PyTorch · Scikit-learn · Text Classification

AI / LLM — OpenAI API · LangChain · LangGraph · CrewAI · Hugging Face · RAG · Agents · Prompt Engineering · Fine-Tuning (LoRA, QLoRA) · Ollama · LiteLLM · Guardrails AI · Embeddings · Semantic Search · Agentic IDEs

Backend & Data — FastAPI · REST APIs · Node.js · PostgreSQL · Redis · Pinecone · pgvector · Microservices · React

Cloud & DevOps — AWS (EC2, S3, Lambda, ECS) · Azure · Docker · GitHub Actions · CI/CD · Git · PyTest · Agile/Scrum


M.S. Computer Science (AI Track), Saint Louis University · B.Tech ECE, JNTU Kakinada

Pinned Loading

  1. aadyon-assist aadyon-assist Public

    Self-hosted, multi-user AI life-ops platform — multi-agent orchestration with human-in-the-loop safety, tiered model routing (LiteLLM), RAG over pgvector. FastAPI + Postgres 16, 7 microservices, 14…

    Python