Everything I'm not makes me everything I am.
I’m a 3rd year CS undergrad and a product-focused engineer from Delhi, India. I have worked in 2x fast-paced startup environments where I have been shipping in light years. I take ownership in 0-1 products and love working with systems, AI and automation.
When not coding, I go for a walk, listen to songs and chill with my friends. I also like to speak about my mind through blogs and X.
I'm open to work, freelance, or collaborate. Contact Me.
Professional Experience
Responenet
Full Stack & AI Intern
December 2025 - Present
- •Reduced infrastructure costs by 96% by replacing Google Maps Directions API with a self-hosted routing engine (Valhalla); built and periodically updated routing tiles using AWS ECS Fargate, stored them on S3, and served them via an EC2
- •Implemented real-time location tracking using Socket.IO (WebSockets) to stream live coordinates between delivery personnel and partners, enabling low-latency tracking and synchronization.
- •Built and integrated an agentic AI assistant using Vercel AI SDK to perform in-chat CRUD operations across the platform; achieved ~95% task accuracy on internal evaluation dataset, improving user interaction efficiency and reducing manual workflows.
- • Led the migration of a 15-year-old WordPress website to Next.js while preserving SEO integrity, including URLs, structured data (JSON-LD), Open Graph, and Twitter metadata, ensuring zero traffic or ranking loss.
- •Designed and implemented a Pub/Sub and Webhook-based automation pipeline to ingest new YouTube uploads from the organization’s channel, automatically syncing videos to the database and rendering them on the website in real time.
Oddmind Innovations
Full Stack AI Engineering Intern
June 2025 - November 2025
- •Designed and implemented a multipart upload pipeline for S3 video recordings, reducing upload time by 90% compared to single-file uploads.
- •Integrated Razorpay Payment Gateway to enable secure and seamless payment processing.
- •Implemented a Docker-based CI/CD pipeline on AWS, every new commit triggers an automated image build, deployment, and EC2 instance update with the latest version.
- •Built a background cron job to batch-send interview feedback reports via Gmail SMTP by periodically querying interview status and ensuring reliable, idempotent delivery.
Chronocept
Dataset Annotator
February 2025 - March 2025
- •Annotated 250+ text samples to train NLP systems on temporal reasoning through segmentation, axis classification, and temporal validity modeling
- •Supported the creation of the Chronocept Dataset, improving AI's understanding of event timelines.
GitHub Contributions ● @21lakshh
Stack I use
Technologies I work with to build products that solve real problems
If you've read this far, you might be interested in what I do.
Let's connect
Find me on these platforms
Here's what I've been listening to lately
WHATEVER WORKS
Kanye West, Ye