# Andrew Wang

![Andrew Wang](/resources/cognition/portraits/andrew-wang.webp)

![Andrew Wang 2](/resources/cognition/portraits/andrew-wang.webp)

MTS at Thinking Machines Lab; ex-Cognition (SWE-1.6) researcher; Terminal-Bench Challenges; MIT  
Thinking Machines Lab  
United States

## At Cognition

- Role: Former Researcher (Research)
- Joined: Jan 2026
- How they joined: Directly
- Now: Thinking Machines Lab (https://www.linkedin.com/in/andrewbwang)

## Identity
- LinkedIn: https://www.linkedin.com/in/andrewbwang
- X: @andr3w_wang
- GitHub: https://github.com/andrewwangva

## Notes
- Currently a Member of Technical Staff at Thinking Machines Lab, a role LinkedIn dates to Aug 2026; that is about four months after his Cognition stint ended. (https://www.linkedin.com/in/andrewbwang)
- On 2026-06-18, between Cognition and Thinking Machines, he posted on X, word for word: "Can agents build complete projects that deliver real value? We’re launching Terminal Bench Challenges: 3 unsolved tasks which could make a real impact on the open source community if solved.

These tasks provide a testing ground for optimizations both on the model and harness level on our continuous leaderboard for each task." The 'We're launching' wording suggests he was on the Terminal-Bench team for this launch. (https://x.com/andr3w_wang/status/2067635236926628133)
- @terminalbench confirmed his authorship on 2026-06-18: "The first three challenges, written by @andr3w_wang, each involve completing projects that would previously require a team of experts working for several months.". (https://x.com/terminalbench/status/2067635275673809337)
- His Cognition window (Jan to Apr 2026) covers the whole SWE-1.6 cycle: Cognition published the SWE-1.6 early preview and research update in early March 2026 and released SWE-1.6 generally in April 2026. (https://cognition.com/blog/swe-1-6)
- LinkedIn lists Cognition (Devin/Windsurf) from Jan 2026 to Apr 2026 with a placeholder title of '.'. (https://www.linkedin.com/in/andrewbwang)
- He follows Adam Zweiger ('Working on learning @thinkymachines | prev: @cognition @MIT_CSAIL'), whose path MIT -> Cognition -> Thinking Machines matches his own. (https://x.com/andr3w_wang)
- In his launch thread he wrote, word for word: "Challenges are tasks that are not only large in size, but also difficult because of the technical ability needed to solve them.

I'm excited to see how these tasks drive harness optimizations, gearing them toward long horizon work. Even more so, I'm excited to see the end artifacts they come up with!" (https://x.com/andr3w_wang/status/2067635240433033354)
- He closed the thread with a call for contributions, word for word: "We’re looking for more challenge ideas! Especially tasks that have interesting verification methods: adversarial testing, fuzzing with LMs, replicating closed source software, etc.

And tasks in specialized domains: Game emulation, FPGAs, GPU driver programming, MoQ, etc.

https://t.co/VG2jd11YXf" The focus on unusual verification methods (adversarial testing, LM fuzzing) points to his research interest in reward and verification design for long-horizon agents. (https://x.com/andr3w_wang/status/2067635242576359882, https://t.co/VG2jd11YXf)
- Chris Barber's Frontier List poll of people who best understand recursive self-improvement (RSI) includes 'Andrew Wang, Thinking Machines' with 1 vote. (https://x.com/chrisbarber/status/2117303283182350422)
- He follows Terminal-Bench and Harbor builders Alex Shaw (founding MTS at Laude Institute), Ryan Marten, Steven Dillmann (Terminal-Bench-Science lead), Ludwig Schmidt and Alex Dimakis, plus @terminalbench and @frontierbench. This supports his involvement in agent-benchmark work. (https://x.com/andr3w_wang)
- Cognition's SWE-1.6 preview post says the model 'learns to use bash commands for search instead of pre-defined tools because terminal commands are more ex[pressive]'. Terminal-native agent behavior is the same theme he worked on next with Terminal-Bench. (https://cognition.com/blog/swe-1-6-preview)
- He follows Cheng-Yuan (Sam) Lee, whose bio reads 'Research @cognition | prev intern @windsurf | 2x ICPC World Finalist'. Lee is the clearest Windsurf-to-Cognition link in his network, but no source shows that Andrew himself came through Windsurf or Codeium. (https://x.com/andr3w_wang)
- The same day he reposted @terminalbench's announcement 'Introducing Terminal-Bench Challenges! A new capability has emerged at the frontier: agents completing large-scale proj[ects]'. His own launch post got 39 likes and 11 reposts. (https://x.com/andr3w_wang/status/2067635409337589781)
- Cognition presented 'How We Scaled RL for Agentic Coding: Lessons from Training SWE-1.6' at NVIDIA GTC San Jose 2026 (session S81825). It is the main public talk on the model family he worked on; the page does not list him as a speaker. (https://www.nvidia.com/en-us/on-demand/session/gtc26-s81825/)
- He follows Thinking Machines researcher Tyler Griggs (@thinkymachines, Berkeley Sky Lab, SkyRL), a current-team connection on RL training infrastructure. (https://x.com/andr3w_wang)
- GitHub account andrewwangva (name 'Andrew Wang', 8 followers) shows no pinned or top-contributed public repositories. His main public open-source-facing work is the Terminal-Bench Challenges launch. (https://github.com/andrewwangva)
- The X account @andr3w_wang (created 2018-12, bio 'MIT', 70 followers, 4 posts) matches his MIT education. He rarely posts; his public posting is launch announcements only. (https://x.com/andr3w_wang)

## Experience
- Member of Technical Staff at Thinking Machines Lab (Aug 2026 - Present)
- . at Cognition (Jan 2026 - Apr 2026)

## Education
- Massachusetts Institute of Technology - Bachelor's degree

## Sources
- https://www.linkedin.com/in/andrewbwang
- https://x.com/andr3w_wang/status/2067635236926628133
- https://cognition.com/blog/swe-1-6
- https://x.com/EdwardSun0909
- https://cognition.com/blog/swe-1-6-preview
- https://www.nvidia.com/en-us/on-demand/session/gtc26-s81825/
- https://x.com/andr3w_wang
- https://x.com/chrisbarber/status/2117303283182350422
- https://github.com/andrewwangva
- https://x.com/terminalbench
- https://www.tbench.ai/contributors
- https://www.tbench.ai/news/terminal-bench-challenges
- https://x.com/terminalbench/status/2067635275673809337
- https://x.com/andr3w_wang/status/2067635240433033354
- https://x.com/andr3w_wang/status/2067635242576359882
- https://t.co/VG2jd11YXf
- https://x.com/andr3w_wang/status/2067635409337589781


Researched with Autumn (https://www.autumn.ai) from public sources. Updated 2026-11-04. Profile: https://www.autumn.ai/resources/cognition/people/andrew-wang · Everyone at Cognition: https://www.autumn.ai/resources/cognition/people