Skip to content

Progress Tracker¶

How to use: Fork or download this file. Check off items as you complete them. Items marked 🌟 are highest-priority for your learning path.

Curriculum Modules¶

Orientation & Foundations¶

🌟 Module 0 — Orientation — Story-first introduction (15 min)
🌟 Module 1 — Introduction — What & Why, prompt anatomy (30 min)
🌟 Module 2 — Core Principles — Specificity, decomposition, iteration, evaluation (45 min)

Patterns & Production Skills¶

🌟 Module 3 — Patterns — Zero-shot, few-shot, CoT, role-playing, ReAct (60 min)
🌟 Module 4 — Best Practices — Token management, versioning, anti-patterns (45 min)

Advanced & Agentic¶

Module 5 — Advanced Patterns — RAG, injection defense, evaluation pipelines (60 min)
Module 6 — Agentic Patterns — Agents, reflection, multi-agent systems (60 min)

Exercises¶

Module 1¶

Exercise 1.1 — Prompt decomposition (five-component rewrite)
Exercise 1.2 — Ambiguity identification
Exercise 1.3 — Comparative analysis (naive vs. engineered)

Module 2¶

Exercise 2.1 — Specificity audit on an existing prompt
Exercise 2.2 — Decomposition design (intra vs. inter-prompt)
Exercise 2.3 — Iteration log (three revision cycles)

Module 3¶

Exercise 3.1 — Pattern identification in production prompts
Exercise 3.2 — Pattern selection for four tasks
Exercise 3.3 — Few-shot vs. zero-shot design comparison

Module 4¶

Exercise 4.1 — Token budget audit
Exercise 4.2 — Prompt refactoring (eliminate duplication across prompt files)
Exercise 4.3 — Anti-pattern identification and fix

Module 5¶

Exercise 5.1 — RAG prompt design for a technical support chatbot
Exercise 5.2 — Red-team your own prompt (three attack types)
Exercise 5.3 — Evaluation pipeline design (5-case test suite)
Exercise 5.4 — Cross-model portability audit

Module 6¶

Exercise 6.1 — Plan-and-execute agent design
Exercise 6.2 — Reflection loop applied to a previous exercise
Exercise 6.3 — Multi-agent code review system design
Exercise 6.4 — Memory management design for multi-session agent

Hands-On Labs¶

🌟 Lab 1 — Zero-Shot vs. Few-Shot (10 min)
🌟 Lab 2 — Chain-of-Thought Impact (10 min)
Lab 3 — Specificity Experiment (10 min)
Lab 4 — Evaluation Pipeline (15 min)
Lab 5 — Tool-Calling & Structured Output (20 min)
Lab 6 — Plan-and-Execute Agent (25 min)
Failure Gallery — Diagnose and fix 5 broken prompts — labs/failure-gallery/ (30 min)

Deep-Dive Comparisons¶

Reference Guides¶

Architecture Decision Records¶

Research Extension Track (~25 hours total)¶

Track 1: Foundations of In-Context Learning¶

Paper 1 — Language Models are Few-Shot Learners [Brown2020]
Paper 2 — Training LMs with Human Feedback [Ouyang2022]

Track 2: Reasoning and Chain-of-Thought¶

Paper 3 — Chain-of-Thought Prompting [Wei2022]
Paper 4 — Large LMs are Zero-Shot Reasoners [Kojima2022]
Paper 5 — Self-Consistency CoT [Wang2023]

Track 3: Agents and Tool Use¶

Paper 6 — ReAct [Yao2023]
Paper 7 — Reflexion [Shinn2023]
Paper 8 — Generative Agents [Park2023]

Track 4: Safety and Robustness¶

Paper 9 — Red Teaming LMs with LMs [Perez2022]
Paper 10 — Indirect Prompt Injection [Greshake2023]

Track 5: Retrieval Augmentation and Evaluation¶

Paper 11 — RAG for Knowledge-Intensive NLP [Lewis2020]
Paper 12 — MT-Bench and Chatbot Arena [Zheng2023]

Track 6: Reasoning Models and Test-Time Compute¶

Paper 13 — Test-Time Compute Scaling [Snell2024]
Paper 14 — Process Supervision / Let's Verify Step by Step [Lightman2023]
Paper 15 — System-2 Attention [Saha2024]

See Research Extension Track for full study guides.

Prompt Templates Used¶

Python Stack¶

React + TypeScript Stack¶

auditor-best-practices.prompt.md
auditor-codebase-maturity.prompt.md
auditor-cybersecurity-features.prompt.md
auto-code-implementation.prompt.md
create-chatbot-ollama.prompt.md
safety-gate-llm.prompt.md

Node.js + TypeScript Stack¶

create-api-endpoint.prompt.md
review-code.prompt.md
write-tests.prompt.md
generate-openapi-spec.prompt.md

Last updated: February 2026. Check CHANGELOG.md for new additions.

← Back to curriculum