FEB / UvA · two-day intensive

End-to-End Research with AI

A hands-on, two-day workshop teaching researchers to work with Claude Code as a research agent — not a chatbot that does the work for you, but a directed collaborator across the full empirical research cycle: literature review, data collection, analysis, and reproduction.

1
Planexpose the decisions before acting
2
Specifyexternalise rules into files, not prompts
3
Superviseverify without writing or reading code
4
Auditreproduce, stress-test, judge credibility

What you'll learn

Four labs, ~2.5 hours each, one research question apiece — together they form a complete empirical research cycle directed entirely from the terminal.

Lab 1 · Specify

The literature review as a pipeline

Appraise a search strategy, turn screening criteria into an executable protocol, validate AI screening against a human gold standard, and catch a model that hallucinates references.

Lab 2 · Plan + Specify

From the web to a dataset

Force planning before collection, externalise a codebook and protocol, run incrementally with provenance on every observation, and diagnose failures at the level of the pipeline, not the row.

Lab 3 · Supervise

Analysis under supervision

Direct an entire analysis without writing or reading code — onboard the agent, separate mechanical from judgment decisions, and resolve a live data crisis purely through supervision.

Lab 4 · Audit

Reproducing and stress-testing a paper

Reverse-engineer a published result, reproduce it independently, bound a defensible specification space, and referee a robustness vs. fragility debate between agents.

Resources

Everything for a pair to get started: the repo, each lab's starter folder as a ready-to-download zip, and every guide, environment file, subagent and skill described in full on the resources page.

Everything

Full package (no GitHub access needed)

The devcontainer, all four labs' starter content, and the reference subagents/skills — one zip, unzip and open in VS Code. Instructor-only material is deliberately excluded.

GitHub

Full source & documentation

All four labs, the infrastructure design notes, and the reference subagents/skills — versioned, browsable, and where updates land first.

Lab 1

The literature review as a pipeline

~440-record corpus, dedup log, gold standard, open-access full texts. Semantic Scholar reference verification runs live via the proxy below.

Lab 2

From the web to a dataset

18 real, curated European companies (7 clear AI-governance / 6 vague / 5 none). The offline mirror is currently the fully-synthetic fallback set — see the resources page for status.

Lab 3

Analysis under supervision

2,500-row synthetic survey engineered so a naive merge silently drops exactly 417 observations — the mid-lab crisis.

Lab 4

Reproducing and stress-testing a paper

A self-contained replication package whose headline result is genuinely fragile — significant without sector controls, gone once you add them.

Infrastructure

Semantic Scholar proxy & site mirror

No participant ever handles an API key or crawls the open web — both run centrally and are described in full on the resources page.

Looking for the six reference subagents, the referee-review skill, or a specific CLAUDE.md? → Full resource index

Download links for installers and tools

Set these up before day one. No Anthropic API key needed — you'll log in to Claude Code with your workshop seat.

Claude Code CLI

The agent you'll direct for all four labs.

curl -fsSL https://claude.ai/install.sh | bash Setup docs →

Docker Desktop

Runs the pinned devcontainer (Python + R + Claude Code, identical for every pair).

Download →

VS Code

With the "Dev Containers" extension, for "Reopen in Container".

Download →

Git

To clone the repo and reset a broken project folder.

Download →

Python 3.11+

Only needed if you're not using the devcontainer at all.

Download →

R 4.x

Only needed if you're not using the devcontainer at all (Lab 4 packages).

Download →

New to devcontainers? → Setup guide: Option A (VS Code + Docker Desktop) and Option C (devcontainer CLI, no VS Code)