Understand the systems behind modern AI. Build with them properly.
The AI Merge helps software engineers build depth in modern AI engineering: models, agents, RAG, vision systems, inference, evals, monitoring, infrastructure, deployment, and AI-assisted engineering workflows.
This GitHub organization is the code side of The AI Merge. It contains reference implementations, walkthrough repos, examples, and companion projects that show how AI components fit into software systems.
These repos focus on the engineering around AI systems, especially the parts that matter when a prototype has to work in a real product:
- how models behave inside applications
- how agents, tools, RAG, and multimodal systems are structured
- how inference, latency, deployment, and infrastructure affect product decisions
- how evals, monitoring, and reliability fit into the system
- how software engineers can use coding agents without losing ownership of the codebase
- End-to-end AI system walkthroughs
- Agents, MCP, A2A, RAG, vision, and multimodal examples
- Inference, deployment, and infrastructure references
- Evals, monitoring, and reliability patterns
- AI-assisted engineering workflows
- Course companion repositories
- Practical examples connected to The AI Merge articles and visual guides
| Area | What it covers |
|---|---|
| AI foundations | Model behavior, tokens, embeddings, attention, fine-tuning, hallucinations, and the fundamentals engineers need before building on top of models. |
| AI systems | RAG, agents, tools, vision, speech, multimodal applications, workflows, APIs, and product architecture. |
| Deployment and infrastructure | Cloud, edge, containers, queues, runtimes, inference engines, GPUs, latency, memory, batching, quantization, and cost tradeoffs. |
| Evaluation and monitoring | Test sets, regressions, traces, feedback loops, confidence gates, observability, and failure analysis. |
| AI-assisted engineering | Specs, coding agents, review loops, TDD, test harnesses, and workflows that keep engineers in control of the codebase. |
The AI Merge is for software engineers who want to understand AI deeply enough to build with it, not just call an API and hope the output holds up.
| Where to start | What you get |
|---|---|
| Website | The main hub for products, courses, visual guides, and current work. |
| Newsletter | Field notes, technical breakdowns, and essays on AI engineering systems. |
| Repositories | Reference code, examples, and implementation patterns. |
These projects are built around complete AI systems, not isolated snippets or toy examples.
| Project | What it is | Status |
|---|---|---|
| The AI Atlas | A 220+ page visual guide to the AI engineering field, with diagrams, mental models, and system maps for modern AI systems. | Join waitlist |
| MAVS | A course on building a full-stack wildlife monitoring AI system, covering MLOps, LLMOps, agents, multimodal workflows, and edge AI inference. | Subscribe for updates |