Backend & AI Systems Engineer · Pune, India

I build systems that think, scale, and barely need babysitting.

4 years of experience building distributed microservices, RAG pipelines, and AI-driven automation. I focus on reducing manual effort through intelligent systems — from LLM-based failure triage to high-throughput data engines.

See my work Get in touch →

About me

Engineer who automates the boring stuff — at scale.

I've spent the last 4 years working on backend systems that handle real load — 30,000+ queries/day, 99.99% uptime, async pipelines moving 12,000+ messages daily. Most recently I've been deep in AI: RAG systems, LLM integrations, and intelligent document processing.

I like building things that reduce toil for engineering teams — whether that's automating failure triage, eliminating manual debugging, or making data actually queryable.

Projects

AI Research Assistant GitHub →

A multi-session research assistant supporting concurrent, memory-aware chats. Integrates OpenAI and Google Gemma 3 for hybrid AI responses, with a multi-agent framework (summarizer, fact-checker) for role-based collaboration.

Next.js FastAPI OpenAI Gemma 3 Multi-agent

Shared Cart — Chrome Extension GitHub →

A Manifest V3 Chrome extension that lets multiple users collaboratively view and add items to a shared cart on any shopping site, in real time. Features a floating UI injected via content scripts, Firebase sync, and secure sandboxed DOM interaction to prevent XSS.

JavaScript Manifest V3 Firebase HTML/CSS

Work experience

Sep 2025
Present

Senior Software Engineer

TEKSystems · HSBC

Full Stack AI Engineer — Advance Data Management AI Solution

Built a distributed AI system across 10 microservices for document processing and workflow automation.
Developed a RAG pipeline to extract and query insights from large PDF datasets, including image-based content processed via PyMuPDF.
Converted formula images to LaTeX using LLMs, improving downstream data usability.
Built an AI-driven failure triage system that analyzes logs and generates structured bug reports with root cause analysis — significantly reducing manual debugging effort.
Designed loosely coupled architecture with centralized authorization and configuration management.

Mar 2024
Aug 2025

Software Engineer

National Payments Corporation of India

Full Stack Engineer — UPI Complaints & Rupay BIN Management

Built and scaled a high-performance query engine serving 30,000 queries/day against data in S3.
Enabled streaming responses to prevent connection timeouts — recovered 50% of previously failing responses.
Integrated Kafka to deliver 12,000+ async responses/day, replacing fragile API-based delivery.
Containerized and deployed in a Docker Swarm microservices architecture, achieving 99.99% uptime.
Designed and built 50+ APIs supporting distributed workflows for the Rupay BIN Management System.
Implemented JWT-based authentication across microservices, improving security and reducing token overhead.

Achievements

2× winner in Coding and Playground Challenges at NPCI.

5-star problem solver on HackerRank in both C++ and Python.

B.E. in Computer Science, SPPU Pune University — CGPA 9.28.