Headline

AI Engineer | Multi-Agent Systems, RAG & LLMs | Full-Stack (React, Node) | Ex-Celsior | Open to Full-Time Roles

About

I'm an AI engineer based in Bangalore. Most recently I spent seven months at Celsior building a multi-agent platform for legacy code modernization: RAG pipelines in LangChain, plus the MCP tools and orchestration logic the agents use to work across brownfield and greenfield codebases. Before that I built full-stack financial applications for COM Financial Services (React, Node, Postgres), and before that I was doing NLP research at NIT Bhopal that turned into a Springer paper on fake news detection. On my own time I built RepoBRD, which takes a GitHub repo and generates a structured 16-section business requirement document by analyzing the actual code rather than guessing from a README. It's the project I'd point to first if you want to see how I think about a problem. I led a team to win the Kavach cybersecurity hackathon in 2023, a 36-hour national event run under the Ministry of Home Affairs, my first time leading a team at that scale. Currently looking for full-time Software Engineer / AI Engineer roles. If you're hiring or know someone who is, I'd like to hear from you.

Location

Bengaluru

AI stack

antigravity

claude code

codex

deepmind

grok

ollama

Work experience

Full Stack Developer at self-emplyed, Sep 2024 - Present · 2 y 1 mo, Bangalore Urban, Karnataka, India, As a freelance software engineer, I design and develop scalable, user-centric web applications and backend systems for startups, small businesses, and personal clients. I specialize in delivering full-stack solutions to transforming ideas into production-ready digital products using modern tech stacks. I collaborate across design and development, ensuring clean code, intuitive UI/UX, and high performance. Projects range from dynamic dashboards and API integrations to machine learning prototypes. Key Contributions: • Engineered full-stack applications using React.js, Next.js, Node.js, and MongoDB. • Designed responsive UI/UX interfaces with Figma and Tailwind CSS. • Deployed scalable APIs and integrated PostgreSQL and Django for backend systems. • Prototyped AI-based features using machine learning and deep learning models.

Artificial Intelligence Intern at Celsior, Jan 2026 - Jul 2026 · 7 mos, Bengaluru, Karnataka, India, I contributed to the development of a multi-agent platform that modernized legacy systems and enhanced automation capabilities. • Built a production-grade multi-agent platform for legacy code modernization, covering both brownfield and greenfield projects • Improved AI task execution and knowledge retrieval by implementing RAG pipelines in LangChain • Accelerated agent automation by building MCP tools and reusable skills for autonomous workflows • Increased platform extensibility by designing modular agent orchestration and communication frameworks

Machine Learning Research Intern at Maulana Azad National Institute of Technology, Nov 2023 - May 2024 · 7 mos, Bhopal, Madhya Pradesh, India, I had the opportunity to contribute significantly to machine learning research at Maulana Azad National Institute of Technology. • Developed NLP and computer vision models in Python, enhancing research methodologies. • Built scalable data preprocessing pipelines, reducing processing time by 40%. • Created a published LSTM-based fake news detection model with an impressive 98% accuracy.

Education

Bachelor of Technology - BTech, Computer Science with Artificial Intelligence and Machine Learning, LNCT Group of Colleges

12th (Intermediate Degree), Physics, Chemistry and Math (PCM), Central Board of Secondary Education

Public links

github: https://github.com/vsuman00

medium: https://medium.com/@vaibhavsuman00

vaibhavsuman.vercel.app.com: https://vaibhavsuman.vercel.app.com/