Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Implement full-text search for AI agent knowledge bases using PostgreSQL tsvector, trigram similarity, and GIN indexes with ranking, fuzzy matching, and hybrid search strategies.
An analysis of emerging patterns in agentic AI and the skills that will remain valuable as the field evolves, with practical advice on where to invest your learning time for long-term career resilience.
Learn how to fuzz test AI agents with automated input generation, boundary testing, adversarial inputs, and crash detection to discover failure modes before users do.
Explore how to leverage Gemini's massive 1 million token context window for analyzing entire codebases, processing full books, and building agents that reason over large document collections.
Master Gemini's function calling capabilities to build agents that use external tools. Learn tool definitions, function declarations, automatic execution, and multi-turn tool use patterns.
Learn how to use Gemini's built-in Google Search grounding to build agents that access real-time information, handle citations properly, and deliver accurate, up-to-date responses.
Build agents that see, hear, and understand multiple media types simultaneously. Learn Gemini's media upload API, inline data handling, video analysis, and audio transcription capabilities.
Implement Gemini streaming for real-time token delivery in agent UIs. Learn stream_generate_content, chunk handling, SSE integration with FastAPI, and building responsive chat interfaces.
Learn how to get reliable JSON output from Gemini using response_mime_type, JSON schemas, enum constraints, and validation. Build agents that produce machine-readable structured data every time.
