Shuvendu Lahiri (@lahirishuvendu) 's Twitter Profile
Shuvendu Lahiri

@lahirishuvendu

Sr Principal Researcher @RISE_MSR, @MSFTResearch. Works in the intersection of programming languages, software engineering and artificial intelligence.

ID: 1425691779962007557

linkhttps://www.microsoft.com/en-us/research/people/shuvendu/ calendar_today12-08-2021 05:33:18

41 Tweet

206 Followers

363 Following

Shuvendu Lahiri (@lahirishuvendu) 's Twitter Profile Photo

Excited to attend FSE 2025 and AIware 2025 to learn about latest research in AI4SE and sharing our work in trusted AI-assisted programming twtr.to/6dau1, including LLM-assisted user-intent-formalization nl2postcond.github.io

Excited to attend <a href="/FSEconf/">FSE 2025</a>  and <a href="/AIwareConf/">AIware 2025</a>  to learn about latest research in AI4SE and sharing our work in trusted AI-assisted programming twtr.to/6dau1, including LLM-assisted user-intent-formalization nl2postcond.github.io
Shuvendu Lahiri (@lahirishuvendu) 's Twitter Profile Photo

Very fortunate to have collaborated with Tom, and also to revive in small parts Tom and (late) Susan's seminal work on program integration/merge conflict resolution through formal methods and ML.

Shuvendu Lahiri (@lahirishuvendu) 's Twitter Profile Photo

Great feedback from attendees at AIware 2025 today (at FSE 2025) presenting Industry perspective on LLM-assisted user-intent-formalization and applications such as verified parser construction from RFC. slideshare.net/slideshow/user… Sarah Fakhry Saikat Chakraborty Nikhil Swamy

Sarah Fakhoury @ ICSE25 (@fakhourysm) 's Twitter Profile Photo

If you're at #FSE2024 talk to Shuvendu Lahiri about our work inferring formal program specs from informal natural language intent. Formal specs can be used for as a mechanism for intent-aligned code generation, or tasks like bug finding! Project page: nl2postcond.github.io

If you're at #FSE2024 talk to <a href="/LahiriShuvendu/">Shuvendu Lahiri</a> about our work inferring formal program specs from informal natural language intent.  Formal specs can be used for as a mechanism for intent-aligned code generation, or tasks like bug finding!
Project page: nl2postcond.github.io
Eddie Aftandilian (@eaftandilian) 's Twitter Profile Photo

I’m #hiring for an ML researcher in GitHub Next. We’re looking for someone with LLM training/fine-tuning experience, strong coding skills, and a passion for AI developer tools to help us shape the future of software development. github.careers/careers-home/j… 1/

Saikat Chakraborty (@saikatch107) 's Twitter Profile Photo

🚨 Thrilled to announce our recently accepted paper at ICSE, publishing benchmarks to push the boundaries of AI-driven Proof Oriented Programming. 🔗 huggingface.co/datasets/micro… 🔗 huggingface.co/datasets/micro… For more details: fstar-lang.org/popai #PoPAI #ICSE2025

Nikolaj Bjorner (@bjornernikolaj) 's Twitter Profile Photo

An internship at MSR, including RiSE, is a great opportunity to work on exciting research projects and mingle with other fantastic students. jobs.careers.microsoft.com/global/en/job/…

Nikolaj Bjorner (@bjornernikolaj) 's Twitter Profile Photo

We, RiSE/MSR, are hiring for exciting and ambitious projects. jobs.careers.microsoft.com/global/en/job/… (on the picture: lunch with colleagues)

We, RiSE/MSR, are hiring for exciting and ambitious projects.

jobs.careers.microsoft.com/global/en/job/…

(on the picture: lunch with colleagues)
Julien Vanegue (@jvanegue) 's Twitter Profile Photo

RiSE/MSR is one of the best places on earth to do PL research. I was blessed to work with this team in a past life and would not be where I am without them today. A++ would recommend. Shuvendu Lahiri Thomas Ball

Sam Blackshear (@b1ackd0g) 's Twitter Profile Photo

Julien Vanegue Shuvendu Lahiri Thomas Ball +1. MSR is awesome, and Shuvendu Lahiri is an amazing researcher and human. I loved my summer working at MSR with Shuvendu, Francesco, and Manuel + there are tons of other great folks there too.

Lakshya A Agrawal (@lakshyaaagrawal) 's Twitter Profile Photo

One of the main goals I had while building out multilspy (aka.ms/multilspy) was that eventually LLMs will be able to tool call LSPs. Happy to see steps in this direction: Checkout MultilspyMCP (playbooks.com/mcp/asimihsan-…), which provides an mcp implementation over multilspy!

Sarah Fakhoury @ ICSE25 (@fakhourysm) 's Twitter Profile Photo

We’ll present our Journal-First work on TiCoder, a workflow for aligning code generation with human intent, producing specs (tests!) along the way. 📖 Test-Driven Interactive Code Generation: User Study and Empirical Evaluation 📆 Friday May 2 📍Hall 1 and 2 ⏰ 12:15

We’ll present our Journal-First work on TiCoder, a workflow for aligning code generation with human intent, producing specs (tests!) along the way.

đź“– Test-Driven Interactive Code Generation: User Study and Empirical Evaluation
📆 Friday May 2
📍Hall 1 and 2
⏰ 12:15
Sarah Fakhoury @ ICSE25 (@fakhourysm) 's Twitter Profile Photo

Finally, don’t miss our talk on Friday about how AI might allow us to achieve verified parsing at scale! 📖 3DGen: AI-Assisted Generation of Provably Correct Binary Format Parsers 📆 Friday May 2 📍Room 212 ⏰ 2:00-2:15

Finally, don’t miss our talk on Friday about how AI might allow us to achieve verified parsing at scale!

đź“– 3DGen: AI-Assisted Generation of Provably Correct Binary Format Parsers
📆 Friday May 2
📍Room 212
⏰ 2:00-2:15