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
Design event-driven AI agent systems using pub/sub messaging, event sourcing for full audit trails, and CQRS to separate read and write models for scalable, observable agent architectures.
Implement event-driven communication between AI agent microservices using Kafka, RabbitMQ, and NATS. Learn event schema design, pub/sub patterns, event sourcing, and exactly-once delivery semantics.
Build resilient event replay infrastructure and dead letter queue management for AI agent systems with proper logging, recovery patterns, and operational tooling in Python.
Implement event sourcing for AI agent systems to create a complete audit trail of every decision, enable workflow replay for debugging, and build projections that reconstruct agent state from event history.
Understand what exascale computing is, why crossing the quintillion-operations-per-second threshold matters, and how these supercomputers accelerate scientific discovery and AI research.
Build production-grade entity extraction with LLMs. Learn schema design for names, dates, addresses, and custom entity types, plus batch extraction techniques and accuracy optimization strategies.
Learn how to combine metadata filters with vector similarity search, comparing pre-filter and post-filter strategies, designing filterable metadata schemas, and building a responsive faceted search UI.
Build automatic failover systems that seamlessly switch between LLM providers when your primary model is unavailable. Learn provider health checks, quality comparison, and cost-aware routing.
Learn how to structure a FastAPI project for AI agent backends, leverage async endpoints for concurrent LLM calls, use dependency injection effectively, and manage application lifecycle with lifespan events.
