Skill Roadmap

AI Agents

AI agents wrap an LLM in a loop — perceive, reason, act, observe — and give it tools, memory, and structure so it can complete multi-step tasks on its own. This roadmap walks you from LLM fundamentals through prompt engineering, tools, memory, agent architectures like ReAct and MCP, and building, evaluating, and securing real agents.

What skills does building AI Agents need?

A layer on top of solid backend and LLM fundamentals

An AI agent is more than a single prompt-response call — it's an LLM wired into a loop that can plan, call tools, remember context, and decide when it's actually done. Building one well means understanding both sides: the LLM itself (tokenization, context windows, generation controls, RAG) and the systems engineering around it (APIs, error handling, state, observability).

Because agents call real tools — code execution, databases, external APIs, even email — security and testing matter as much as capability. Prompt injection, tool sandboxing, and evaluation pipelines aren't optional extras; they're what separates a demo from something you'd trust in production.

Most teams also lean on a framework (LangChain, CrewAI, LangGraph) rather than building the agent loop from scratch every time — but understanding what those frameworks are doing under the hood is what lets you debug them when things go wrong.

The AI Agents Roadmap

Work through these in order — LLM fundamentals first, then the agent layer on top

STEP 1

LLM Fundamentals

What's actually happening inside the model you're building on top of.

Transformer Models & LLMs Tokenization Context Windows Token Based Pricing Open vs Closed Weight Models Model Families & Licences Streamed vs Unstreamed Responses Reasoning vs Standard Models Fine-tuning vs Prompt Engineering Embeddings & Vector Search Basics of RAG
STEP 2

Generation Controls

The parameters that shape how an LLM's output is actually sampled.

Temperature Top-p Frequency Penalty Presence Penalty Max Length Stopping Criteria
STEP 3

AI Agents 101

The core loop that turns a language model into something that can act.

What are AI Agents? What are Tools?
The Agent Loop
1. Perception / User Input 2. Reason and Plan 3. Acting / Tool Invocation 4. Observation & Reflection
Example Use Cases
Personal Assistant Code Generation Data Analysis Web Scraping / Crawling NPC / Game AI
STEP 4

Prompt Engineering

Getting reliable behavior out of a model before you ever add tools.

Be Specific in What You Want Provide Additional Context Use Relevant Technical Terms Use Examples in Your Prompt Specify Length, Format, etc. Iterate and Test Your Prompts
STEP 5

Tools / Actions

Giving the model a way to actually do things beyond generating text.

Tool Definition
Name and Description Input / Output Schema Error Handling Usage Examples
Examples of Tools
Web Search Code Execution / REPL Database Queries API Requests Email / Slack / SMS File System Access
STEP 6

Agent Memory

Letting an agent carry context across a conversation, or across sessions entirely.

What is Agent Memory?
Short Term vs Long Term
Within Prompt Vector DB / SQL / Custom Episodic vs Semantic Memory
Maintaining Memory
RAG & Vector Databases User Profile Storage Summarization / Compression Forgetting / Aging Strategies
STEP 7

Agent Architectures

Established patterns for structuring how an agent reasons and coordinates tools.

ReAct (Reason + Act) RAG Agent Planner / Executor DAG Agents
Model Context Protocol (MCP)
MCP Hosts MCP Client MCP Servers Core Components Creating MCP Servers
Deployment Modes
Local Desktop Remote / Cloud
STEP 8

Building Agents

From raw API calls to reaching for a framework once the pattern gets familiar.

Manual (From Scratch)
Direct LLM API Calls Implementing the Agent Loop Parsing Model Output Error & Rate-limit Handling
LLM-Native Function Calling
OpenAI Function Calling Gemini Function Calling OpenAI Assistant API Anthropic Tool Use
Building Using Frameworks
STEP 9

Evaluation and Testing

Proving an agent actually does what it's supposed to, reliably.

Metrics to Track Unit Testing for Individual Tools Integration Testing for Flows Human in the Loop Evaluation
Evaluation Frameworks
STEP 10

Debugging and Monitoring

Seeing what an agent actually did, once it's more than a single prompt.

Structured Logging & Tracing
Observability Tools
STEP 11

Security & Ethics

An agent that can act in the world needs guardrails around what it's allowed to do.

Prompt Injection / Jailbreaks Tool Sandboxing / Permissioning
STEP 12

Self-Critique and Multi-Agent Systems

Scaling from one agent to several that check each other's work or specialize by role.

Self-critique Agents Multi-Agents

GitHub Projects

Real repos to practise the exact workflows in this roadmap

Frequently Asked Questions

Common questions from people starting out with AI Agents

What's the difference between an AI agent and a chatbot?

A chatbot mostly turns one input into one output — a reply to a message. An agent runs a loop: it can decide to call a tool, look at the result, and decide what to do next, repeating until it judges the task done, rather than just generating a single response.

How is an agent different from plain RAG?

Retrieval-Augmented Generation fetches relevant context once and hands it to the model to answer with. An agent can use RAG as one tool among several, but it also decides when to retrieve, when to call something else instead, and when it needs another round of reasoning before answering.

What is the Model Context Protocol (MCP)?

MCP is an open standard for connecting an AI agent (the host/client) to external tools and data sources (servers) through a consistent interface, so a tool built once can be reused across different agent applications instead of writing custom integration code for each one.

Do I need a machine learning background to build agents?

Not really — most agent development is calling existing LLM APIs and building the software around them, which leans more on backend engineering than on training models. A conceptual understanding of how LLMs generate text is enough to get productive.

Should I use a framework like LangChain, or build from scratch?

Building the agent loop manually once is worth doing to understand what's actually happening. For anything beyond a toy project, a framework handles tool schemas, memory, and error handling for you — but debugging framework behavior is much easier once you've built the loop yourself.

How do I prepare for an AI agent engineering interview?

Be able to explain the agent loop and tool-calling in your own words, understand the tradeoffs between short-term and long-term memory, know how you'd evaluate whether an agent actually succeeded at a task, and be ready to discuss prompt injection and how you'd sandbox a tool with real side effects.

Track complete

From a single prompt to a tool-using, memory-aware agent you can actually evaluate and secure — that's the core of what building AI agents involves today. Keep experimenting with real tools and real failure cases; that's where the interesting engineering happens.

Where next?

Keep exploring by domain or drill into a single skill