ResearchCraft

Enrolling now · Live online workshop

AI-Assisted Research Workflows

A 3-hour hands-on workshop on using AI agents across the research pipeline — literature discovery, code generation, experiment iteration, and modular paper writing — while keeping every claim and citation verifiable.

Duration
3 Hours
Format
Online Workshop
Modules
8

Curriculum

8 modules, one session

3 Hours of live teaching, in the order you'll use it on a real project.

  1. 01

    The Agentic Research Loop

    What AI agents can and can't be trusted with across the research pipeline, and where a human checkpoint is non-negotiable.

  2. 02

    Literature Discovery Agents

    Building an agentic search-and-summarise pipeline over arXiv, Semantic Scholar, and PubMed with citation verification.

  3. 03

    AI Pair-Programming for Research Code

    Using Claude Code and GitHub Copilot to write, test, and debug experiment code faster.

  4. 04

    Automating Experiment Iteration

    Agentic loops for hyperparameter sweeps, logging, and result triage.

  5. 05

    Modular AI-Assisted Paper Writing

    Drafting each IMRaD section with AI assistance, then assembling and voice-checking the full paper.

  6. 06

    Verifying AI Output

    Catching hallucinated citations, fabricated numbers, and overclaimed results before they reach a reviewer.

  7. 07

    Prompting & Tool Use for Research

    Prompt patterns, MCP tools, and connecting agents to your own data and code safely.

  8. 08

    Disclosure & Journal AI Policies

    What top journals and conferences require you to disclose about AI use, and how to stay compliant.

Outcomes

What you'll leave with

01

Agentic Pipeline

A working literature-discovery and summarisation pipeline you can reuse.

02

Verification Checklist

A checklist for catching hallucinated citations and claims before submission.

03

AI-Assisted Draft

A drafted paper section produced and verified with an agentic workflow.

04

Disclosure Statement

A compliant AI-use disclosure for your target venue.

Tools & frameworks you'll use

  • Claude Agents
  • OpenAI API
  • Claude Code
  • LangChain
  • MCP
  • Elicit
  • Scispace
  • GitHub Copilot
  • Automation Pipelines
  • Vector DB

Enroll

Ready to build your own research agents?

Pay securely by card and your seat is confirmed instantly — we'll email you the joining details.

  • Full refund if you cancel before the start date
  • Small cohort — 15 to 25 people

Keep going

More live courses

See all courses