Your AI expertise path

Ship-quality Python for a JavaScript-fluent engineer — the delta from JS plus AI-app patterns: Python idioms and typing, async/await and FastAPI, testing and clean project structure.

• Python for JavaScript DevelopersNot started
• Async & APIs with FastAPINot started
• Testing & Project StructureNot started

How LLMs work (transformers, attention, tokenization) and how to steer them: generation parameters, calling LLM APIs, prompt and context engineering, provider/model selection, and fine-tuning basics.

• How LLMs WorkNot started
• Generation & DecodingNot started
• Calling LLM APIsNot started
• Prompt EngineeringNot started
• Context EngineeringNot started
• Provider & Model SelectionNot started
• Fine-Tuning BasicsNot started

Retrieval-augmented generation end to end: embeddings, chunking and indexing, vector databases, retrieval quality (hybrid search, reranking), and RAG evaluation. The worked-example track.

• RAG FundamentalsNot started
• Chunking & Indexing StrategiesNot started
• Vector DatabasesNot started
• Retrieval QualityNot started
• RAG EvaluationNot started
• RAG Frameworks — LangChain & LlamaIndexNot started

The through-line of the whole market: building tool-using agents. ReAct loops, tool/function calling, LangChain and LangGraph, MCP, memory and multi-step design, multi-agent patterns, and code-vs-no-code judgment.

• Agent FundamentalsNot started
• Tool Use & Function CallingNot started
• LangChain EssentialsNot started
• LangGraph OrchestrationNot started
• MCP — Model Context ProtocolNot started
• Memory & Multi-step DesignNot started
• Multi-agent PatternsNot started
• Code vs No-code JudgmentNot started

The "beyond the model" differentiators: eval design and metrics, hallucination and reliability, observability and tracing, and the LLMOps lifecycle (versioning, reproducibility, drift, incident management).

• Eval DesignNot started
• Hallucination & ReliabilityNot started
• ObservabilityNot started
• LLMOps LifecycleNot started

Guardrails and content safety (prompt shielding, injection defense), access and data protection (RBAC, PII, tenant isolation), and governance / responsible AI (compliance, EU AI Act awareness, bias, transparency).

• Guardrails & Content SafetyNot started
• Access & Data ProtectionNot started
• Governance & Responsible AINot started

Shipping to production: containerization (Docker), CI/CD for AI apps, the Azure AI platform (Azure OpenAI, Azure AI Search, Foundry / Agent Framework) as primary, orchestration at scale (Kubernetes, IaC), and AWS/GCP equivalents cross-referenced.

• ContainerizationNot started
• CI/CD for AI AppsNot started
• Azure AI PlatformNot started
• Orchestration at ScaleNot started
• Cloud Equivalents (AWS & GCP)Not started

The 8/8 skill the market always asks for: turning AI into business value. Discovery and scoping, ROI and success criteria, stakeholder translation and adoption, culminating in a capstone — shipping an end-to-end agent as portfolio evidence.

• Discovery & ScopingNot started
• Business Value & ROINot started
• Stakeholder TranslationNot started
• Capstone: Ship an End-to-End AgentNot started

The forward-deployed / AI-GTM flavor: REST and webhooks, event-driven workflows (queues, async pipelines, retries), and business-tool connectors (CRM, ticketing, data hubs).

• REST & WebhooksNot started
• Event-Driven WorkflowsNot started
• Business-Tool ConnectorsNot started