# Agents in Research

Progress in using AI agents to empower researchers is rapid but still very unevenly distributed. Many researchers have been reporting great benefits while others are still complaining about ChatGPT hallucinating references. This is a collection of public accounts by researchers who shared their usage and lessons learned. It was compiled by AI agents at the direction of [Dominik Lukeš](https://dominiklukes.net) who also provided some of his own use cases.

## Lessons

1. [Choose a question worth answering](/lessons/choose-a-question-worth-answering/index.md): Before asking an agent to tackle a research problem, ask who would care about the answer and what it would tell them. Revisit those questions when the first results come in. A calculation can become feasible long after specialists know what answer to expect.
2. [Build the method and its checks together](/lessons/build-the-method-and-its-checks/index.md): An agent can help build the tools a research method needs. Decide while building how you will tell whether the method works. For a new search, compare what it returns with existing searches. For a simulation you want to repeat, review the setup before reusing it on new inputs.
3. [Use a council of models](/lessons/use-judgement-where-it-changes-the-work/index.md): When I have a decision that matters, I give several models the same question and brief and ask them to answer separately. I compare their reasons, especially where they disagree, before deciding what to do.
4. [Start with a known result and keep an independent check](/lessons/start-with-a-known-result/index.md): Try the method first on a result you already know. Before moving on to new work, decide how you will check an answer when there is no published result to compare it with. Where possible, make that check independent of the code that produces the answer.
5. [Keep enough of the work to resume, audit and undo it](/lessons/keep-enough-to-resume-and-undo/index.md): Keep the original material, the last usable output and a record of what failed. If a long run stops or a correction goes wrong, you need to know what you can reuse and what you have to do again. A saved decision can also keep a new session from repeating a failed approach.
6. [Routine research work is a useful place to start](/lessons/start-with-routine-research-work/index.md): Try an agent on a recurring task whose output you can check against the project files, such as preparing for a meeting from past notes or flagging inconsistencies in a manuscript. A task you do often gives you a reason to spend time explaining it and a chance to see whether the next attempt takes less work to check.

## Accounts

- 20 Nov 2026: [John H Boyer](/accounts/boyer-greek-readers-agentic-coding/index.md), philosophy. [Source](https://johnhboyer.com/)
- 20 Nov 2026: [Thomas D. Grant](/accounts/grant-using-llms-in-scholarship-international-law/index.md), law. [Source](https://www.arsinq.com/inferpedia/static/arsinquirendi/slides/grant-using-llms-in-scholarship-2026.pdf)
- 27 Oct 2026: [Exploring a question in knot theory (Robert Lipshitz)](/accounts/lipshitz-khovanov-local-equivalence/index.md), mathematics. [Source](https://arxiv.org/html/2610.37595v1)
- 24 Oct 2026: [Tracing summaries to a radio archive (Dominik Lukeš)](/accounts/dominik-lukes-audioarchive/index.md), digital archiving. [Source](https://audioarchive.semanticmachines.fyi/)
- 1 Oct 2026: [Using physics tools to study a forest (Matthew Schwartz)](/accounts/schwartz-claude-shaped-science/index.md), physics. [Source](https://www.anthropic.com/research/claude-shaped-science)
- Oct 2026: [Chris Blattman](/accounts/blattman-claude-blattman-research-management/index.md), economics. [Source](https://claudeblattman.com/how-i-use-ai/)
- 25 Sep 2026: [William Baude](/accounts/baude-claude-code-law-review-article-experiment/index.md), law. [Source](https://blog.dividedargument.com/p/claude-generated-legal-scholarship)
- 18 Sep 2026: [Lawrence Evalyn](/accounts/evalyn-claude-code-subscription-lists-transcription/index.md), literary studies. [Source](https://newsletter.dancohen.org/archive/reader-responses-summer-2026)
- 13 Aug 2026: [Simon Goldstein](/accounts/goldstein-epistocracy-philosophy-paper/index.md), philosophy. [Source](https://dailynous.com/2026/08/13/philosophy-journal-publishes-largely-ai-authored-article-on-purpose-guest-post/)
- 13 Aug 2026: [Terence Tao](/accounts/tao-sendov-lean-formalization/index.md), mathematics. [Source](https://github.com/teorth/sendov/blob/master/docs/making-of.md)
- 11 Aug 2026: [Gus Skorburg](/accounts/skorburg-ai-for-philosophy-cowork/index.md), philosophy. [Source](https://gus1365199.substack.com/p/how-i-use-ai-for-philosophy)
- 21 Jul 2026: [Checking an R port against Stata (Francis DiTraglia)](/accounts/ditraglia-ivreg2r-stata-validation/index.md), economics. [Source](https://restatr.com/ivreg2r/articles/validation.html)
- 15 Jul 2026: [Ke Xiao, Han Li, Teng Zhang, Yangchen Xu, Runze Mao, Zhi X. Chen](/accounts/xiao-cfd-case-bundle-codex/index.md), engineering. [Source](https://arxiv.org/abs/2609.11941)
- Jul 2026: [Robert Lipshitz](/accounts/lipshitz-claude-code-working-mathematician/index.md), mathematics. [Source](https://pages.uoregon.edu/lipshitz/projects/claude/)
- 24 Jun 2026: [Pedro H. C. Sant'Anna](/accounts/santanna-claude-code-my-workflow-academic-lifecycle/index.md), economics. [Source](https://psantanna.com/claude-code-my-workflow/)
- 24 Jun 2026: [Aaron Tay](/accounts/tay-agentic-library-catalogue-search-mcp/index.md), library science. [Source](https://aarontay.substack.com/p/what-changes-when-an-llm-agent-searches)
- 12 Jun 2026: [Scott Cunningham](/accounts/cunningham-claude-code-economics-data-verification/index.md), economics. [Source](https://www.aieconomist.io/articles/scott-cunningham-claude-code-ai-economists)
- 9 Jun 2026: [Adam Johnson](/accounts/johnson-llm-assisted-omop-code/index.md), medicine. [Source](https://www.ohdsi.org/wp-content/uploads/2026/06/LLM-OHDSI-Pres-Johnson-6-9-26.pdf)
- 20 Apr 2026: [Joshua Sternfeld](/accounts/sternfeld-claude-code-mcp-chronicling-america/index.md), history. [Source](https://encodingthepast.substack.com/p/on-ai-enabled-research-historic-newspapers)
- 8 Apr 2026: [Jia Yu, Weiwei Yu, Pengfei Xiao, Fukun Xing](/accounts/yu-agent-driven-corpus-linguistics/index.md), linguistics. [Source](https://arxiv.org/abs/2604.07189)
- 2 Apr 2026: [Seyed Yahya Shirazi](/accounts/shirazi-neuroscience-research-skills/index.md), neuroscience. [Source](https://neuromechanist.github.io/blog/010-agentic-research-workflows)
- 29 Mar 2026: [Paul Goldsmith-Pinkham](/accounts/goldsmith-pinkham-claude-code-applied-economists/index.md), economics. [Source](https://markusacademy.substack.com/p/claude-code-for-applied-economists)
- 27 Mar 2026: [Cameron Blevins](/accounts/blevins-mail-transit-history/index.md), history. [Source](https://cblevins.github.io/posts/bottleneck-side-quests)
- 23 Mar 2026: [Siddharth Mishra-Sharma](/accounts/mishra-sharma-differentiable-boltzmann-solver/index.md), physics. [Source](https://www.anthropic.com/research/long-running-claude)
- 23 Mar 2026: [Matthew Schwartz](/accounts/schwartz-vibe-physics-sudakov-shoulder/index.md), physics. [Source](https://www.anthropic.com/research/vibe-physics)
- 16 Mar 2026: [Max Zimmer, Nico Pelleriti, Christophe Roux, Sebastian Pokutta](/accounts/zimmer-pokutta-agentic-researcher/index.md), computer science. [Source](https://arxiv.org/abs/2603.15914)
- 5 Mar 2026: [Seth Lazar](/accounts/lazar-minty-agent-philosophy-lab/index.md), philosophy. [Source](https://philosophyofcomputing.substack.com/p/how-to-use-coding-agents-for-philosophy)
- 24 Feb 2026: [Zeke Hausfather](/accounts/hausfather-ai-augmented-scientist/index.md), climate science. [Source](https://www.theclimatebrink.com/p/the-ai-augmented-scientist)
- 20 Feb 2026: [Sam Illingworth](/accounts/illingworth-claude-code-paper-folder-peer-review-skill/index.md), higher education. [Source](https://theslowai.substack.com/p/ai-agents-changing-academic-research)
- 19 Feb 2026: [Benjamin J. Shields](/accounts/shields-cmxflow-agentic-cheminformatics/index.md), chemistry. [Source](https://b-shields.github.io/posts/2026/02/cmxflow)
- 15 Feb 2026: [Rob Patro](/accounts/patro-cpp-to-rust-genomics-tools/index.md), genomics. [Source](https://combine-lab.github.io/blog/2026/02/15/a-skeptics-guide-to-generative-ai-coding.html)
- 23 Jan 2026: [Tom Pepinsky](/accounts/pepinsky-agentic-ai-social-science-practice/index.md), political science. [Source](https://tompepinsky.com/2026/01/23/agentic-ai-and-social-science-research-practice/)
- 19 Jan 2026: [Rayed AlGhamdi](/accounts/alghamdi-claude-code-thematic-analysis-alignment/index.md), information systems. [Source](https://doi.org/10.1177/16094069261462093)
- 14 Jan 2026: [Shawn Graham](/accounts/graham-claude-code-newspaper-ocr-jack-the-ripper/index.md), history. [Source](https://electricarchaeology.ca/2026/01/14/futzing-with-newspaper-ocr/)
- 13 Jan 2026: [Tim Dettmers](/accounts/dettmers-use-agents-or-be-left-behind/index.md), computer science. [Source](https://timdettmers.com/2026/01/13/use-agents-or-be-left-behind/)
- 13 Jan 2026: [Andy Hall](/accounts/hall-vote-by-mail-replication/index.md), political science. [Source](https://freesystems.substack.com/p/the-100x-research-institution)
- 1 Jan 2026: [Koen van Gilst](/accounts/vangilst-un-general-assembly-debates-claude-code/index.md), digital humanities. [Source](https://un.koenvangilst.nl/research)
- 9 Jul 2025: [Kevin Yang](/accounts/yang-claude-code-data-analysis/index.md), field not stated. [Source](https://yang3kc.substack.com/p/claude-code-is-secretly-an-excellent)
- 4 Mar 2025: [José R. Penadés](/accounts/penades-ai-co-scientist-phage-hypothesis/index.md), microbiology. [Source](https://medios.uchceu.es/actualidad-ceu/en/professor-jose-r-penades-tests-google-co-scientist-system/)
- 2026: [Matthias Mittner](/accounts/mittner-psychopy-nback-ai-coding/index.md), psychology. [Source](https://ihrke.github.io/posts/psychopy_ai_coding.html)
- 2026: [Tiago Ventura](/accounts/ventura-coding-agents-political-science/index.md), political science. [Source](https://tiagoventura.github.io/talks/agentic_ai/claude_faculty_retreat.html)

Also: [replication attempts](/replications/index.md), [key papers](/papers/index.md), [guides and tools](/resources/index.md), [about](/about/index.md).
