Architecting high-performance backend systems, optimizing complex databases (SQL/NoSQL/Aerospike), and building universal developer tooling. Passionate about open-source innovation, clean concurrency, and scalable engineering.
- 🚀 High-Performance Backend Systems: Specializing in low-latency microservices, concurrent execution models in Go & C++, and efficient memory management. Re-architected request handling at Zeta using a Command Pattern execution model, reducing EOD batch processing time by 60% (from 4 hours to 1.5 hours).
- 🔍 Advanced Database Engineering: Deep expertise in relational (PostgreSQL, MySQL, OracleDB) and NoSQL (MongoDB, Redis, Aerospike) optimization, dynamic indexing, and complex schema querying.
- 🤖 AI & LLM Systems: Architecting privacy-first AI workflows using local LLMs (Ollama) and speech-to-text models (OpenAI Whisper) to ensure zero data leakage in sensitive domains.
- 🛠️ Universal Developer Tooling & GitOps: Passionate about reducing developer friction through cross-platform SDK management, automated dependency resolution, AWS Lambda serverless execution, and Kubernetes reconciliation loops.
| Project & Overview | Architecture & Tech Stack |
|---|---|
| ⚙️ GitopsCtl - GitOps Reconciler CLI A lightweight GitOps CLI tool implementing reconciliation loops to sync Kubernetes manifests with Git state. |
Highlights: Application registration, kubeconfig management, real-time sync monitoring. Stack: Go, Kubernetes |
| 🌉 AI IDE Bridge (llm-bridge) A universal proxy and adapter enabling developers to use any AI IDE's model catalog (Cursor, Copilot, Windsurf) from any OpenAI-compatible client. |
Highlights: Multi-provider OAuth flows, MCP (Model Context Protocol) integration, universal LLM gateway CLI. Stack: TypeScript, Node.js, MCP |
| 🎙️ AshaVoice - LLM Medical Scribe A privacy-first voice-to-text assistant that transcribes and summarizes doctor-patient conversations into structured SOAP notes. |
Highlights: Locally hosted LLMs (Ollama) ensuring zero data leakage and compliance with sensitive medical workflows. Stack: Python, Docker, OpenAI Whisper, Ollama |
| 📦 Universal SDK Manager (USDK) A robust tool to manage, version, and integrate SDKs across multiple programming platforms (Python, JS, PHP) effortlessly. |
Highlights: Unified SDK interface, automated dependency resolution, AWS Lambda serverless execution. Stack: Go, Gin, PostgreSQL, AWS Lambda |
- Zeta (Jun 2025 - Present) | SDE 1: Designed and built end-to-end credit card statement-to-loan conversion system. Re-architected request handling using a Command Pattern execution model, reducing EOD batch processing time by 60% (from 4 hours to 1.5 hours). Led AWS multi-zone cluster setups.
- Hatio / Billdesk (Nov 2023 - May 2025) | SDE 1: Built a high-performance UPI transaction switch (P2M) using Java (Spring Boot) and microservices, improving throughput and reducing settlement latency. Developed a secure bulk payout processing system, reducing errors by 20%.
- Give India (2023) | SDE Intern: Improved CSR fund management systems and enhanced user workflows.
- eGovernments Foundation (2022 - 2023) | SDE Intern: Contributed to hospital management dashboards for digital healthcare transformation.
"I'm always open to collaborating on complex backend systems, developer tooling, and open-source initiatives. If you find my projects valuable or want to discuss systems architecture, make sure to follow my GitHub and reach out!"
