Software Engineer with 3+ years of backend and full-stack experience building high-throughput distributed systems at scale. At Farmart, architected an audit trail platform, a multi-channel notification microservice processing 2K+ daily notifications, and a dynamic image resizing service that cut asset sizes by over 50% — while reducing cloud costs by 30%.
Previously built and owned critical microservices at Livlong 365 processing 15–20K orders/day. IIT Delhi graduate with a strong foundation in systems design, open-source tooling, and AI-adjacent engineering including MCP servers and LLM integrations.
Farmart Service Pvt. Ltd.
Gurugram, India
Architected and built a resilient audit trail platform for FarmartOS using a transactional outbox pattern, Node.js/Sequelize middleware, JSONB payload schemas, transaction-linked multi-entity events, idempotent retry-safe consumers, and custom ESLint rules to enforce audit-event creation inside database transactions. Engineered a scalable multi-channel communication microservice using Node.js and AWS SQS to orchestrate 2K+ notifications/day across Push (FCM), SMS, and WhatsApp; built the FCM engine with MongoDB-backed rate limiting, dynamic instance management, and automated token purging for reliable delivery at scale.
Built a Dynamic Image Resizing Service using AWS CloudFront and Lambda@Edge, enabling on-the-fly resizing of image assets and reducing their size by more than 50%. Led backend development for the B2C app (8K+ DAU) and platforms team; optimized critical API response times using database indexing, Redis caching, query optimization, and event-driven design; containerized applications with Docker and deployed as right-sized AWS ECS tasks with horizontal auto-scaling.
Reduced cloud costs by 30% through AWS infrastructure ownership and right-sizing; improved interservice authentication security by implementing rotating API keys. Built a centralized logging library using OpenTelemetry collectors as a sidecar across 10 services, exporting logs to OpenSearch data streams; enriched traces via Data Prepper and reported errors using Sentry.
Livlong 365 Wellness and Insurance (Backed by IIFL)
Mumbai, India
Developed and maintained backend systems across 12 microservices using .NET, MSSQL, and MongoDB, implementing Web APIs, JWT authentication, Redis caching, Azure Service Bus, Elasticsearch, rate-limiting middleware, and third-party API integrations. Took ownership of critical services including Authentication, Product, Orders, ShortURL, and a PDF generation service; the Orders service processed 15–20K orders/day; developed a custom CSV parser from scratch and spearheaded adoption of a cleaner onion architecture.
Built internal and client-facing portals in Angular and React serving 1–2K DAU; resolved critical security vulnerabilities identified in VAPT audits and converted HTML/CSS to PDF using iTextSharp. Awarded Emerging Performer in the technology department (2022–23).
Bachelor of Technology
Delhi, India
Node.js, TypeScript, Model Context Protocol, Outline API
Built an open-source Model Context Protocol (MCP) server in TypeScript that enables AI assistants like Claude to read and write documents in Outline, a popular team knowledge base tool. Implemented tools for full-text search, document CRUD operations, and collection management by wrapping the Outline REST API — allowing AI models to interact with the knowledge base through natural language.
Compatible with all major MCP clients including Claude, Cursor, and GitHub Copilot; gained 15 GitHub stars from the community, demonstrating adoption across AI-powered development environments.
TypeScript, Node.js, Express, MongoDB, Next.js, Figma Plugin API
Built and open-sourced a Figma plugin that auto-fills text layers with realistic mock data using a layer-naming convention (e.g. a layer named mobile number gets a real-looking number) — adopted by the Far Mart design team for Farmart OS prototypes, with a backend designed to support any team’s custom datasets.
Architected a full-stack system with a Figma Plugin API frontend, Express + MongoDB backend, and Next.js management dashboard with Google OAuth 2.0; worked around the Figma sandbox’s network restrictions via postMessage between the plugin UI iframe and sandbox, with datasets fetched live so additions are available without any plugin update.
TypeScript, VS Code Extension API
Built a VS Code extension in TypeScript that automatically detects and validates CDN URLs and asset links (images, fonts, videos, etc.) across 10+ file types including JS, TS, HTML, CSS, Vue, and Svelte. Implemented non-blocking background validation using lightweight HTTP HEAD requests with 5-minute smart caching to avoid redundant network calls, and debouncing to prevent scanning on every keystroke.
Broken links are surfaced as inline warnings directly in the editor’s Problems panel, supporting CDN providers such as CloudFront, Cloudflare, Akamai, Fastly, jsDelivr, and unpkg.
Secured All India Rank - 1926
Cleared KVPY Stage 1 examination conducted by IISc Bangalore
Selected for a summer research internship at the University of Oxford and Rutherford Appleton Laboratory under Prof. Peter Norreys and Prof. Rajeev Pattathil (cancelled due to COVID-19).