AI Agent Roadmap
Papers
Initializing search
ai-agent-roadmap
Home
Roadmap
Stages
Reference
AI Agent Roadmap
ai-agent-roadmap
Home
Roadmap
Stages
Stages
Roadmap.sh Alignment
00 Orientation
00 Orientation
What Are AI Agents
01 Prerequisites
01 Prerequisites
Backend Basics
Git and Terminal
02 LLM Fundamentals
02 LLM Fundamentals
Transformer Models
Tokenization
Context Windows
Model Families and Licenses
Generation Controls
Pricing and Latency
Embeddings and Vector Search
RAG Basics
QAs
Junior Interview
Senior Interview
03 Prompt Engineering
03 Prompt Engineering
Prompt Basics
Writing Good Prompts
Structured Outputs
Prompt Testing
QAs
Junior Interview
Senior Interview
04 Agent Fundamentals
04 Agent Fundamentals
Agent Loop
ReAct Pattern
Reasoning and Planning
Acting and Observation
Stopping Criteria
Tools Overview
QAs
Junior Interview
Senior Interview
05 Tools and Actions
05 Tools and Actions
Tool Definition
Tool Schemas
Function Calling
Tool Error Handling
Common Agent Tools
Boundaries and Destructive Tools
QAs
Junior Interview
Senior Interview
06 MCP
06 MCP
MCP Overview
MCP Hosts, Clients, Servers
Building MCP Servers
Local vs Remote MCP
Tool and Resource Exposure
Security Boundaries for MCP-Connected Tools
QAs
Junior Interview
Senior Interview
07 RAG and Memory
07 RAG and Memory
Embeddings and Vector Search
Vector Databases, SQL, and Custom Stores
Short-Term and Long-Term Memory
Episodic and Semantic Memory
User Profile Storage
Summarization, Compression, and Forgetting
Additional Explanation
08 Agent Architectures
08 Agent Architectures
ReAct Agents
RAG Agents
Routing and Prompt Chaining
Planner-Executor Workflows
Evaluator-Optimizer Loops
DAG Agents and Tree-of-Thought Patterns
When a Simpler Workflow Is Better Than an Agent
09 Building Agents
09 Building Agents
Building an Agent Loop From Scratch
Direct LLM API Calls
Parsing Model Output
Error and Rate-Limit Handling
Provider-Native Function Calling and Tool Use
Agent Frameworks
Choosing a Framework Without Losing Debuggability
10 Multi-Agent Systems
10 Multi-Agent Systems
Supervisor-Worker Pattern
Handoffs
Agents As Tools
Shared State vs Message Passing
A2A Protocol
Coordination Failure Modes
How To Tell When One Agent Is Enough
11 Evaluation and Observability
11 Evaluation and Observability
Agent Quality Metrics
Tool Unit Tests
Flow Integration Tests
Prompt Regression Tests
RAG Evaluation
Tracing and Structured Logging
Human Review
Evaluation and Observability Tools
12 Security and Ethics
12 Security and Ethics
Prompt Injection and Jailbreaks
Insecure Output Handling
Excessive Agency
Tool Sandboxing and Permissions
Data Privacy and PII Redaction
Bias, Toxicity, and Guardrails
Safety and Red-Team Testing
OWASP LLM Top 10
13 Production Deployment
13 Production Deployment
Production Architecture
API Design
Background Jobs and Queues
Docker
CI/CD
Monitoring and Alerts
Cost Control
Rollbacks and Incident Response
Reference
Reference
Glossary
Resources
Best Practices
Papers
Add curated papers for AI-agent learning here.
Back to top