Suvansh Sanjeev (@SuvanshSanjeev) / X

Suvansh Sanjeev

577 posts

Suvansh Sanjeev

@SuvanshSanjeev

Research @OpenAI. ex-🤖 @berkeley_ai, @CMU_Robotics. Teamwork makes the dream work, and Dreamworks makes Shrek.

Joined February 2016

Pinned
Suvansh Sanjeev
@SuvanshSanjeev
Aug 7, 2025
GPT-5 is what you’ve been waiting for – it defines and extends the cost-intelligence frontier across model sizes today. it’s been a long journey, and we’ve landed pivotal improvements across many axes in the whole GPT-5 family. and hey no more model picker (by default)!
20K
Suvansh Sanjeev
@SuvanshSanjeev
Jun 17, 2025
i left my phd before joining openai working in industry demands more rigor – you don’t just need to convince reviewer 2 with a nice graph and an ego-cite, it better actually work if it’s underwriting billions in research investment not saying it always pans out that way in
Bob McGrew
@bobmcgrewai
Jun 17, 2025
Many of OpenAI's greatest researchers did not have PhDs in AI. Building a path for brilliant technical people without experience to become researchers was critical for our success.
434K
Suvansh Sanjeev
@SuvanshSanjeev
May 24, 2023
Don't have access to ChatGPT Code Interpreter? Me neither. But with Open Data Interpreter @BrilliantlyAI, you can chat w/your CSVs—free! 📊 Ask for graphs, tables, and more: datainterpreter.brilliantly.ai Built on @langchain, made possible by support from @trychroma! (vid sped up)
00:00
133K
Suvansh Sanjeev
@SuvanshSanjeev
May 30, 2024
i started as a researcher at @OpenAI last month on @BorisMPower's team 💨 to help make models useful in new domains oh also signed a non-capitalization agreement 🤷🏽
68K
Suvansh Sanjeev
@SuvanshSanjeev
Apr 28, 2023
Use the power of GPT-4 to chat with @nflfastR! 🤖🏈 I’ve built Blitz blitz.brilliantly.ai to answer your burning @NFL stats/#NFLDraft queries 💭 How do your fave QBs stack up? 💭 Heaviest player w/ TD pass in the past 5 years? (Spoiler: @tkelce) 💭 Lightest 1st round QBs?
40K
Suvansh Sanjeev
@SuvanshSanjeev
Jan 8, 2021
Let’s PaVE the way forward for NFL passing analytics with #BigDataBowl! Introducing Passing Value in Expectation (PaVE)! Joint work with the wonderful team of @rishavd64 @uditranasaria @aditya_murali7. kaggle.com/dutta64/pave-t… Thread:
00:00
Suvansh Sanjeev
@SuvanshSanjeev
Apr 14, 2025
we trained a cute lil model! - will solve a few AIME problems a year if you ask nicely - cheap image input, 1 million token context window - comparable intelligence / instruction-following to 4o-mini - cheapest model we've ever released ($0.12/MTok blended)
Kevin Lu
@_kevinlu
Apr 14, 2025
Today we released GPT-4.1 nano, an amazing effort led by @johnohallman and @SuvanshSanjeev! Some cool features of today's release: - Faster & cheaper than 4o-mini - Significantly cheaper for image processing - Better reasoning across the board - 1M input context
21K
Suvansh Sanjeev
@SuvanshSanjeev
Aug 5, 2025
this was a first-class effort worked on by amazing researchers, and the results speak for themselves I'm proud of OpenAI for this release – open-weights models are huge for Broadly Distributing the Benefits of AI research glad this model made it out alive 🙃
OpenAI
@OpenAI
Aug 5, 2025
Our open models are here. Both of them. openai.com/open-models
11K
Suvansh Sanjeev
@SuvanshSanjeev
Aug 16, 2024
Armen’s been a hacker at heart since we were kids pulled up to PE class one day freshman year with his laptop running his ASR system that could issue computer-navigating actions. couple months later he won a Kaggle competition detecting Alzheimer’s from brain scans. a couple
Armen Aghajanyan
@ArmenAgha
Aug 16, 2024
I've left FAIR/Meta, it's time to build.
14K
Suvansh Sanjeev
@SuvanshSanjeev
Jul 27, 2023
🦾🤖 Say hi to CLAIRe, your newest AI friend! The next step to human-AI conversation dynamics with multi-resolution memory, active messaging, and more: brilliantly.ai/blog/claire Open source, built on @trychroma + @langchain, and made as part of Chroma's hacker-in-residence
brilliantly.ai
CLAIRe: Conversational Learning AI with Recall
Experience a new paradigm of AI conversation. CLAIRe responds, remembers, and reaches out.
25K
Suvansh Sanjeev
@SuvanshSanjeev
Aug 7, 2025
GPT-5 Nano is the cheapest reasoning model offered by any lab – $0.05/$0.40. It’s incredibly intelligence dense, with over 54% SWE-Bench and nearly 10% FrontierMath on "high" reasoning. No model of any size had reached either mark as recently as January – it beats o1, DeepSeek
12K
Suvansh Sanjeev
@SuvanshSanjeev
Sep 12, 2024
o1-preview, the last new generation of models to fall victim to 9.11 vs 9.9
This post is unavailable.
23K
Suvansh Sanjeev
@SuvanshSanjeev
Apr 7, 2023
Meta's Segment Anything Model is exciting and the demo was incredibly well-executed, but many were let down by the teased-and-withheld natural language prompting component. I spent the better part of yesterday playing around with SAM and I'm eager to share a consolation prize!
00:00
12K
Suvansh Sanjeev
@SuvanshSanjeev
Sep 12, 2024
Replying to @SuvanshSanjeev
in fairness, this is an interesting answer
7.4K