- Joined
- Nov 2023
- How they joined
- Directly
- Before Cognition
- quadric.ioSoftware Engineer
- Olympiads
- IOI silver 2019IOI 2019Silver
- Adhyyan Sekhsaria is identified as a Founding Engineer at Cognition.
- Adhyyan Sekhsaria said (verbatim, translated from Chinese): million-level context lets them feed in a complete diff at once, achieving higher-quality code review with a simpler architecture.Mar 14, 2026
- Adhyyan Sekhsaria was an undergraduate computer science student at USC Viterbi and part of a three-person team with master's student Jeel Tejaskumar Vaishnav and undergraduate Kangmin Tan ('22).Aug 1, 2022
- Adhyyan Sekhsaria's team placed among the top 20 schools in the ICPC North American Championship and in the top three teams in the Western region, advancing to the final level of the competition.Aug 1, 2022
- ICPC 2020 World Finals Moscow: rank 81, team University of Southern California (USC Trojans), score 3/15.Oct 5, 2021
- IOI 2019: rank 78, silver medal, score 332.93/600.Aug 11, 2019
- Adhyyan Sekhsaria competed for India at the 2019 International Olympiad in Informatics, scoring 332.93 (55.49%), ranking 78/327 absolute (76.45% relative), and winning a Silver medal.2019
- Sekhsaria quote: "If you're an engineer, the number one thing to think about is being obsessed with customers and making sure they're having a good experience using your product"
- Sekhsaria quote on the key metric: "The main metric we care about is the percent of users who go from the landing page to merging a PR through Devin" and "That metric doubled."
- Adhyyan Sekhsaria's team solved 6 problems, landing 12th place overall amongst 50 teams at the ICPC North American Championship.Aug 1, 2022
- ICPC 2020 World Finals teammates: Kangmin Tan and Tianyi Chen.Oct 5, 2021
- Olympiad: Silver medal in IOI 2019 at Azerbaijan amongst 500,000 students across the world.2019
- ML Compilers - High Performance Library for NN inference
A git repo completely operated by the agent is much more powerful than a single file on your local machine. It handles merge conflicts, observability and scales to much more information. Every cloud agent should have this primitive and that's why we've open sourced the spec. It's lightweight and straightforward to incorporate in any harness!
@maolbac It should just work! You can see your memory in Customize -> Memory
We've spent the past few months iterating on the nuances of this internally. Initially we saw memories degrading Devin runs because they used to store over-generalized learnings. Dreaming was able to fix a lot of this!
Devin magic
devin was already cracked. now its also stacked!
Another day in the trenches with @cognition https://t.co/e2XGy3CZhW
on the path to making agents feel more human! https://t.co/BU6IsbRy8d
Underrated feature about Devin review: the UI doesn't lag for big PRs
Jev + prompt caching would cut costs by 10x more Too cheap too meter!
@tamarajtran @typesafeai best jev use case i've seen so far!
ppl dont realize that fable is now cheaper than opus
HUGEEE
