# neural-maze/agentic-patterns-course

Implementing the 4 agentic patterns from scratch

Repository: https://github.com/neural-maze/agentic-patterns-course
Canonical: https://ross.abutalabs.com/products/agentic-patterns-course
Language: Jupyter Notebook
License: MIT
License Family: permissive
Last push: 2025-03-18T16:37:23+00:00

## Health v2 (maintenance only)
Score: 31/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 12, release rhythm 40, longevity 56
- inputs: {"age_days": 784, "days_push": 533, "days_rel": 677, "gap_med": 4, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1743, forks 336 (observed 2026-08-28T04:05:30.450687+00:00)

## What it is
A Jupyter Notebook course implementing the four agentic patterns (reflection, tool use, planning/ReAct, and multiagent) from scratch using raw Groq API calls, with no frameworks like LangChain or CrewAI. It includes step-by-step notebooks and complete Python implementations for each pattern.

## Use cases
- learn how to build AI agents from scratch without frameworks
- implement the reflection pattern for LLM self-improvement
- build a ReAct agent with tool use using raw API calls
- understand Andrew Ng's four agentic patterns
- create a multiagent crew with plain Groq API calls
- study agent design patterns with worked notebooks

## When to choose
- you want to understand agent internals rather than use LangChain or CrewAI abstractions
- you prefer learning via step-by-step Jupyter notebooks
- you want framework-free implementations using the Groq API

## When to avoid
- you need a production-ready agent framework with built-in features
- you use a different LLM provider and don't want to adapt the Groq-specific code
- you need maintained, versioned software rather than educational material

## Facets
- artifact type: learning-resource
- maturity: active
- function: agent-framework, llm-inference, prompt-engineering
- domain: artificial-intelligence, large-language-models, tutorials
- platform: python
- tags: agentic-patterns, groq, jupyter-notebooks, reflection-pattern, tool-use, planning-pattern, multiagent, from-scratch, educational, ai-agents

## Member repositories
- neural-maze/agentic-patterns-course (main) score 31

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:30.450687+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:29:34.005199+00:00, confidence not recorded.
  - readme: https://github.com/neural-maze/agentic-patterns-course (fetched 2026-08-28T04:05:30.450687+00:00, sha 966b89288833)
- Data as of 2026-08-30T08:39:29.467469+00:00.