I’m a software engineer with a wide range of skills in the industry.
I'm pretty much head over heels for algorithms and cloud computing. From the moment I wrote my first line of code during my undergrad years in Bengaluru, India, it was like a spark lit up. Now, as a Master's student at Arizona State University, I'm all in—studying AI, machine learning, and everything in between.
At work, I'm the guy who lives for that eureka moment when solving complex problems. Take my time at PhonePe, for example. I was a key part in developing backend features that not only made the user experience smoother but also caught a monumental bug that saved the company a cool 10 million. Trust me, there's nothing like the adrenaline rush of watching your code click into place and realizing you've just saved the day (and a ton of money).
And now at AWS Bedrock, I get to play at an even bigger scale—building infrastructure that powers large language models across hundreds of clusters. There’s something incredibly satisfying about tuning systems that handle millions of requests and shaving seconds off latency, knowing that performance gains at this scale truly move the needle.
When I'm off the clock, you might find me diving into a game of badminton or exploring new hiking trails. I've also mentored over 50 up-and-coming techies, which has been its own kind of thrill. So, yeah, I'm a bit of a Renaissance man, constantly juggling my love for tech, my knack for problem-solving, and my zest for life. What about you? What gets you out of bed in the morning?
M.S in Computer Science
B.E in Information Science
SDE, Amazon Web Services - Bedrock
At AWS Bedrock, I work on the infrastructure that powers large-scale LLM inference across hundreds of Kubernetes clusters. My work focuses on reliability, performance, and capacity orchestration for production AI workloads handling tens of millions of requests per day. I redesigned the cluster capacity rebalance workflow to prevent resource contention during concurrent operations, improving success rates from 60% to over 95%. I also built weighted routing mechanisms using consistent hashing to improve cache locality, reducing model prefill latency from ~20 seconds to ~6 seconds under heavy load. Beyond performance, I’ve led initiatives to productionize new clusters globally—automating validation, rollout, and monitoring workflows that reduced launch time from nearly a day to just a few hours. I’ve also worked extensively on observability and trace export systems to help teams debug inference behavior safely at scale. The work sits at the intersection of distributed systems, cloud infrastructure, and AI workloads—building platforms that enable model teams to ship faster without compromising reliability.
SDE Intern, Amazon - Fire TV
During my time at Amazon's FireTV division on the Display Ads team, I tackled a particularly tricky issue: partners who sponsor campaigns sometimes upload incorrect data. If left unchecked, these errors could cause the system to drop key elements, leading to incomplete campaigns being delivered to users' devices. To resolve this, I spearheaded the development of an end-to-end campaign verification dashboard. Built with a mix of React, Java, and cloud technologies, this dashboard provides real-time previews and verification features, allowing stake holders to catch and correct errors before they impact the campaign. The result? A more reliable and efficient system that ensures only fully-vetted campaigns make it to the screen, enhancing the user experience and fostering better decision-making among our partners.
Software Engineer - Backend, PhonePe
As for my time at PhonePe, it was nothing short of a rollercoaster ride. Working on backend systems, I was the invisible hand making sure your transactions were smooth and secure. From helping you find stores that are actually open when you need them, to saving the company a whopping 10 million INR through smart regulation implementation, my code touched multiple facets of the business. I was also the go-to guy for handling product bugs during high-traffic events like the Indian Premier League. So if you've ever made a last-minute bet without a glitch, you're welcome!
SDE Intern, Bro4u.com
During my time at Bro4U, I tackled some of those everyday headaches that nobody likes to deal with. You know how paperwork is always a mess? I streamlined it. Now, instead of field agents slogging through IDs, a system sorts it all in a snap. And customer feedback? I made it so we can quickly gauge if people are loving the service or if we need to step up our game. It's like I gave the company a pair of reading glasses, making it easier to focus on what really matters.
Programming
Distributed Systems & ML Infra
Backend & Data
Cloud & DevOps
Contact me
Wanna chat? Just shoot an email and I'll get back to you within 24 hours - unless I'm off on some wild adventure or deep in a Netflix binge. Either way, I promise to hit you back as soon as I can!