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 service-to-service authentication for multi-agent architectures using mTLS and service tokens. Covers certificate setup, trust boundaries, token exchange, and FastAPI integration.
Learn how to build secure session management for AI agent conversations. Covers session token design, server-side storage, expiration, concurrent session handling, and forced invalidation with FastAPI.
Design shared memory architectures for multi-agent teams that enable collective knowledge building, with contribution tracking, conflict resolution, and access control.
Master shared state management in multi-agent systems using the OpenAI Agents SDK's RunContext, including shared context objects, state mutation patterns, race conditions, and consistency strategies.
Implement the sidecar pattern to add consistent observability to AI agent microservices without modifying application code. Learn Envoy proxy configuration, log collection, and metric export.
Master SIP trunking to connect AI agents to PBX systems and the PSTN. Covers SIP protocol basics, trunk configuration, DTMF handling, and intelligent call routing for enterprise telephony.
Build a production-ready Slack bot agent using the Slack SDK that listens to events, handles slash commands, responds with interactive messages, and integrates LLM-powered reasoning for team support.
Learn how to implement slot filling patterns in conversational AI agents that collect required information through natural, multi-turn dialog instead of rigid form-like interactions.
Learn how to design a model routing system that sends simple queries to cheap models and escalates complex ones to powerful models. Reduce AI agent costs by 40-60% while maintaining quality with intelligent routing.
