Skip to content

Latest commit

 

History

1 Commit

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AI Workflows

Practical, student-oriented workflows for using AI responsibly in research, software, and academic writing.

These guides treat AI as a fast junior assistant — not as the author of your project, code, or paper. You stay responsible for correctness, understanding, and academic integrity.

Author: Gerald Schuller, TU Ilmenau
(with assistance from ChatGPT for drafting and structuring)


What's in this repository

Guide Description
AI-Assisted Project Workflow End-to-end workflow for research or software projects: problem definition → research → planning → implementation → verification → writing → integrity checklist
AI-Assisted Workflow for Writing Software Incremental software development with AI: specs, structure, small steps, testing, debugging, quality, and docs
AI-Assisted Workflow for Reading & Evaluating Research Papers How to read and critically evaluate papers with AI as a discussion partner — not a substitute for reading
AI-Assisted Workflow for Checking Your Own Paper or Thesis Pre-submission self-check: structure, clarity, technical correctness, reviewer simulation, consistency, references
Structure for a Scientific Engineering Paper or Thesis Recommended structure for engineering papers/theses (problem → method → results → meaning)

Matching PDF versions of most guides are included in this repository for easy sharing and printing.


Guiding principles

  1. Human first — Define the problem, read the abstract, or re-read your draft before you open an AI tool.
  2. Understand everything — If you cannot explain the code, math, or argument, you are not done.
  3. Verify always — Check citations, equations, results, and claims with independent sources and your own judgment.
  4. Disclose when required — Follow your university’s AI policies and document AI use if required.
  5. Iterate with tests — Especially for software: formulate requirements and verification criteria, then iterate until satisfied. AI may write drafts; you own correctness.

Good uses of AI

  • Brainstorming and outlining
  • Explaining concepts, APIs, and equations
  • Debugging and suggesting tests
  • Improving clarity and structure of text
  • Simulating a critical reviewer

Bad uses of AI

  • Replacing your own thinking or reading
  • Submitting unverified AI output
  • Fabricating or accepting fake citations
  • Copying code or text you do not understand

Simple mental models

Context Think of AI as… Not as…
Project / software A fast junior assistant The project author
Reading papers A discussion partner A substitute for thinking
Your own paper A fast pre-reviewer The author of your research

How to use these guides

  1. Pick the workflow that matches your current task.
  2. Follow the steps in order — early “human-only” steps matter.
  3. Adapt prompts and checklists to your course, lab, or university rules.
  4. Use the final integrity checklist before any submission.

License

This work is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0).

You may share and adapt the materials for any purpose, including teaching, as long as you give appropriate credit to Gerald Schuller / TU Ilmenau.


Related note (test-driven AI use)

A short companion idea from class practice:

  • Precisely formulate what you want and how you will test/verify it.
  • Give that to the AI (or a class chatbot).
  • Iterate until you are satisfied.
  • Document requirements, testing, verification, and results.
  • AI might write — you are responsible for correctness.

About

Practical, student-oriented workflows for using AI responsibly in research, software, and academic writing.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors