Wenduo Cheng

A lot of scientific know-how already exists in Gtihub repos, APIs, notebooks, docs, and research papers. But agents still cannot really make use of it out of the box because it is scattered everywhere, We built SkillFoundry to bridge that gap. It turns fragmented scientific resources into reusable #skills that agents can actually use. The basic idea is to use a Domain Knowledge Tree to guide the search, mine candidate skills from heterogeneous resources, package them into executable skills, test them automatically, and then keep refining the library based on what works, what fails, and what overlaps. With the agent skills automatically designed by SkillFoundry, we see gains on 5/6 MoSciBench datasets, and Codex + SkillFoundry does much better on cell annotation than Codex alone, while staying competitive with systems like SpatialAgent. We also gave Biomni automatically designed skills from SkillFoundry for the scDRS workflow, and it outperformed Biomni running on its own. Project: #paper" target="_blank" rel="nofollow noopener">ma-compbio-lab.github.io/SkillFoundry/#… Paper: arxiv.org/abs/2604.03964 Thanks to my co-authors @WenduoC @mishamamq @TurcanAlistair @martinjzhang and @jmuiuc for guidance and support throughout this work.

We are grateful to @NatureComms for featuring our #DNALONGBENCH as one of the #Editors #Highlights in #ComputationalBiology: nature.com/collections/cf…

Happy to share #DNALONGBENCH @NatureComms. We challenge DNA #FoundationModels w/ long-range sequence context and hope this sparks more meaningful ways to evaluate growing AI+BIO models. Kudos to @WenduoC @ZhenqiaoSong @zocean636. Great collab w/ @lileics. nature.com/articles/s4146…

Final version of our L2G paper is now published at @TmlrOrg! Kudos to @WenduoC for leading this work. Paper: openreview.net/forum?id=5NM4g…

Happy to see a time series workshop @ NeurIPS 2025 motivated in part by our search for BERT moments in specialized foundation models (🧵here: x.com/khodakmoments/…)

Excited to present our work at #ICLR2025! Join us on Thursday, April 23 (Afternoon Session) to chat about the current state of foundation models in specialised domains 🧬📈 Hall3+2B #40 Joint work with @ZongzheX2001 @WenduoC @musishen @JunhongShen1 @atalwalkar @khodakmoments

John Zimmerman presenting at #CHI2025 Catch him and Ask about our AI Literacy paper — best paper honorable mention @meslami2 We thank @d19fe8 @koedinger @amyogan @HanZiyuh @du_yanlin for their feedback on this journey. programs.sigchi.org/chi/2025/progr…

I have some news to share! @datadoghq is forming a new AI research lab, and I'm excited to announce that I've joined as Chief Scientist to lead this effort. Datadog has a great work culture, lots of data and compute, and is committed to open science and open sourcing. Our team is working on ambitious research areas grounded in real-world challenges in cloud observability and security, with three current areas of focus: 1. Observability Foundation Models for forecasting, anomaly detection, and multi-modal telemetry analysis (logs, metrics, traces, etc.). 2. Site Reliability Engineering (SRE) Agents to detect, diagnose, and resolve production incidents. 3. Production Code Repair Agents that leverage code, logs, and runtime data to identify and fix performance issues. On a personal note, I'm thrilled to work with Oli and Alexis again (we worked together in the early aughts before they co-founded Datadog). I’m also excited that Datadog, as part of its expansion in AI, is partnering with CMU (note: I will continue to work part-time at CMU and maintain my research activities). Datadog is actively hiring out of our NYC office!

Proper and meaningful benchmark datasets are crucial for advancing genomic LLMs/FMs, and ML methods for genomics in general. Fantastic collab w/ @lileics's group. Amazing work led by @WenduoC @ZhenqiaoSong @zocean636

Can we skip genomic Foundation Model pretraining? Our work L2G repurposes language LLMs for genomics via cross-modal transfer, matching fine-tuned genomic FMs. Kudos to @WenduoC & amazing collab w/ @atalwalkar. L2G, language to genome; L2G, life’s too good biorxiv.org/content/10.110…

🧵 on surprising revelations from our study of specialized foundation models (FMs beyond vision/text): after evaluating dozens of scientific & time series FMs we found that most weren’t even competitive with simple supervised models, some with as little as 513 parameters. 1/n

I'm sharing my @UCLA_CGSI talk slides last week surveying recent #LLM methods in genomics (DNA, scRNA). As exciting as LLM's potential in genomics is, a note of skepticism remains. Hope we maintain vigilance in what we publish. Feedback on slides welcome. cs.cmu.edu/~jianma/talks/…