← Learning path / Sources & credits
Transparency
Sources and credits.
We build on the work of outstanding educators, researchers and open-source communities. This page records every source we use, how we use it, and the licence that applies.
How we reference
1 · In the lesson
Where an explanation is inspired by a source, the lesson says so and links to it.
2 · Continue exploring
Every lesson ends with resources that explain the same ideas from different angles.
3 · This register
The author, link, how we use it and the licence, for every source.
What our labels mean
- Linked
- We point learners to the original. Nothing is copied.
- Referenced
- We cite specific chapters or papers as further reading.
- Inspired by
- The source shaped our teaching approach; our text, diagrams and code are original.
- Adapted from
- We modified the source's material, under its licence, with required notices. Currently: none.
- Reproduced
- We used the source's material unchanged, with permission or under its licence. Currently: none.
We never present another educator's work as our own. All lesson text, diagrams, interactive labs, examples and quizzes on these pages are original Code With Impact material. Example organisations, people, documents and studies are fictional.
Register of sources
| Source | Creator | How we use it | Licence | Lessons |
|---|---|---|---|---|
| LLM Visualization (opens in a new tab) | Brendan Bycroft | Linked. Inspired our step-by-step approach to explaining a model. | Software: MIT | 02 |
| Transformer Explainer (opens in a new tab) | Polo Club of Data Science, Georgia Tech | Linked. Inspired the “type a sentence and follow it” design of our Lesson 02 lab. | Software: MIT | 02 |
| The Illustrated GPT-2 (opens in a new tab) | Jay Alammar | Linked. Inspired our one-concept-at-a-time visual style. | © the author | 02 |
| Attention in Transformers (opens in a new tab) | 3Blue1Brown (Grant Sanderson) | Linked as further reading. | © the creator | 02 |
| Hands-On Large Language Models (opens in a new tab) | Jay Alammar & Maarten Grootendorst | Referenced: chapters cited as further reading. | Book © O'Reilly; companion code Apache 2.0 | 01–05 |
| An Illustrated Guide to AI Agents (opens in a new tab) | Maarten Grootendorst & Jay Alammar | Referenced: chapters cited as further reading. | Book © O'Reilly; companion code has no licence stated, so we link only | 02, 06–07 |
| AI Engineering (opens in a new tab) | Chip Huyen | Referenced: chapters cited as further reading. | © O'Reilly | 01, 03–07 |
| Agentic Design Patterns (opens in a new tab) | Antonio Gullí | Referenced: our selection of agent patterns follows its chapters; all explanations and examples are our own. | © Springer | 03, 05–07 |
| Building effective agents (opens in a new tab) | Erik Schluntz & Barry Zhang, Anthropic | Referenced. Our “autonomy ladder” draws on its distinction between workflows and agents; wording and visuals are our own. | © Anthropic | 06 |
| RAG Techniques (opens in a new tab) | Nir Diamant | Linked only. | Custom non-commercial licence; no code reused | 05 |
| Generative AI for Beginners; AI Agents for Beginners; RAG Time (opens in a new tab) | Microsoft | Linked. | MIT | 01, 05, 06 |
| LLM Course; Agents Course (opens in a new tab) | Hugging Face | Linked. | Apache 2.0 | 01, 04, 06 |
| Short courses (opens in a new tab) | DeepLearning.AI | Linked. | Provider's terms | 01, 03, 07 |
| Prompt Engineering Guide (opens in a new tab) | DAIR.AI | Linked. | MIT | 03 |
| OpenAI Cookbook; Claude Cookbooks (opens in a new tab) | OpenAI; Anthropic | Linked. | MIT | 03, 07 |
| Ragas; DeepEval (opens in a new tab) | Ragas; Confident AI | Linked. Named in Lesson 07's “Go deeper” section. | Apache 2.0 | 07 |
| Chroma; LlamaIndex; Ollama; Streamlit; LangGraph 101 (opens in a new tab) | Various | Linked; small original code examples show typical usage. | Apache 2.0 / MIT | 04–07 |
| Guidance on AI and data protection (opens in a new tab) | Information Commissioner's Office | Linked and referenced. Our summary of data protection principles draws on it. | Open Government Licence | 08 |
| Data protection impact assessments (opens in a new tab) | Information Commissioner's Office | Linked as further reading. | Open Government Licence | 08 |
| Charities and artificial intelligence (opens in a new tab) | Charity Commission | Linked and referenced for trustee responsibilities and internal AI policies. | Open Government Licence | 08–11 |
| AI Playbook for the UK Government (opens in a new tab) | Government Digital Service | Linked and referenced for adoption principles. | Open Government Licence | 09–12 |
| AI Risk Management Framework (opens in a new tab) | NIST | Linked and referenced for risk governance and measurement. | Public domain (US government work) | 09–12 |
| Model risk management principles (SS1/23) (opens in a new tab) | Bank of England, PRA | Referenced for model inventory, independent validation and monitoring. | Bank of England terms | 12 |
Research papers cited
- Vaswani et al. (2017), “Attention Is All You Need” (opens in a new tab), Lesson 02
- Mikolov et al. (2013), “Efficient Estimation of Word Representations in Vector Space” (opens in a new tab), Lesson 04
- Reimers & Gurevych (2019), “Sentence-BERT” (opens in a new tab), Lesson 04
- Lewis et al. (2020), “Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks” (opens in a new tab), Lesson 05
- Yao et al. (2022), “ReAct: Synergizing Reasoning and Acting in Language Models” (opens in a new tab), Lesson 06
Spotted a missing credit or an error? Please tell us and we'll put it right. Last reviewed: October 2026.