[ Portfolio 2026 ]
Software engineer
Backend & distributed systems
I build scalable backend systems: event-driven services, real-time sync and networks that keep working when the internet doesn't.
Software Engineering Intern at DragonX Systems
I like to build. Backend systems that stay fast, consistent and honest under load: event-driven services, real-time collaboration and mesh networks that keep talking when the towers go down. Two thousand solved problems taught me to think in trade-offs. Merged pull requests in production open source taught me to ship them.
Problems solved
Peak LeetCode rating
Merged pull requests
Open-source projects
Backend systems for offline mesh messaging, real-time collaboration, distributed chat and fast URL redirection. Each one is a case study.
Also built
Every repository on GitHubA Raft-based key-value store with leader election, log replication, snapshotting and recovery.
Java · Raft
A RAG-based search engine to search, explore and understand Indian court judgements.
Java · RAG
High-concurrency ticket booking built for flash-sale traffic, with Redis reservations, Kafka and Saga compensation.
Java · Kafka · Redis
An AI-powered maritime solution that won Techfest Hackathon 2026.
Hackathon winner
Internship · On-site / Hybrid
Software Engineering Intern
Jul 2026 – Present
Building GPU-accelerated ASIC physical-design infrastructure: CUDA components integrated into the OpenLane and OpenROAD EDA toolchain.
Faster GCD timing repair
Faster Ibex GPU timing repair
Education
B.Tech, Computer Science
2023 – 2027
Merit-based scholarship with a full tuition waiver for the top admission rank.
Fixes and features in projects that run in production around the world, from Gradle's build tooling and the Apache RocketMQ broker to LINE's Armeria framework.
Merged PRs
Projects
It started with DSA and contests. Thousands of problems later, choosing the right data structure under a time limit is a reflex.
Problems solved across platforms
Peak LeetCode rating, Knight, top 2%
Peak Codeforces rating, Specialist
Rated contests
Core stack: Java, Spring Boot, Kafka, Redis, PostgreSQL, Docker, C++, CUDA, WebSockets, Kubernetes, MySQL
A systematic way into hard problems, each habit grounded in a project it shaped.
Break the problem into its core components. Find the edge cases, the constraints and the scale it has to survive, and ask the clarifying questions before writing a line of code.
Real example
ConvoMesh separates connection handling, routing, caching and persistence, so message delivery scales without tight coupling.
Assume the network, the node or the process will die, and decide up front what the system should say when it does. Honest states beat optimistic ones.
Real example
ResQNet persists every envelope before acknowledging it, and reports an undelivered message as expired, never as delivered.
Partition the work so any node can be added or lost. Route by key, share state through a broker and keep services stateless wherever possible.
Real example
The Sync Engine routes every document through a consistent hash ring to one owner node, and keeps three nodes in step over Redis Pub/Sub.
Treat the log as the source of truth. When every change is an event, any node can rebuild exactly what it lost.
Real example
Kafka event sourcing plus snapshots: load snapshot 100, replay events 101 to 104, and the document is back at version 104.
Benchmark before believing anything. Profile, find the real bottleneck, change one thing, then measure again.
Real example
Profiling-led CUDA work at DragonX Systems made GCD timing repair 15.7× faster, and load tests held 118 ms average latency at 500 concurrent editors.
Smart-Sagar, an AI-powered maritime solution.
Max rating 2102, in the top 2% of contestants.
Max rating 1434 across rated rounds.
22 merged pull requests across Gradle, Spring Kafka, Apache RocketMQ, Armeria, OpenRewrite and more.
Full tuition waiver for the top admission rank at Galgotias College.
Ran a Model United Nations conference for 200+ delegates, Nov to Dec 2025.
Currently interning at DragonX Systems and open to backend, distributed systems and infrastructure roles. Have a role, a system or a hard problem? Send it over.
Or write directly