Vibeleaderboard
← Back to Vibers
alex zhang
Builder

alex zhang

@a1zhang

3 Intel

Alex L. Zhang is a PhD student at MIT CSAIL, advised by Omar Khattab and Tim Kraska, working on language model efficiency and applications, including recursive language models and compositional generalization. He graduated from Princeton (2024), his department's top student, and is an NSF Graduate Research Fellowship recipient. He has held research roles at Sakana AI, VantAI, Snap Research, and Apple, created benchmarks including VideoGameBench and SWE-bench Multimodal, and helps lead the GPU MODE community.

Is this you? Sign in with X to claim this profile.

Intel

Speculative Programmatic Tool Calling(alexzhang13.github.io)

Technique for pre-launching tool calls from partially generated REPL code to hide latency.

Developer ToolsAgents & Harnessesbuilt by @a1zhang
Academia is for Ambition — Alex Zhang, MIT(www.latent.space)

Latent Space interviews MIT's Alex Zhang on Recursive Language Models, context offloading, subagent swarms, KernelBench and GPU Mode, and treating harnesses as compositional generalizers.

AI Agents
Recursive Language Models — Alex Zhang, MIT PhD(www.youtube.com)

Alex Zhang, the MIT researcher behind Recursive Language Models, discusses harnesses as compositional generalizers, programmatic subagent calling, agent swarms, AI-written GPU kernels and capability overhang in current models.

AI Agents