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 multi-agent systems that span organizational boundaries with proper API contracts, trust boundaries, data sharing controls, and compliance frameworks. Build federated agent architectures safely.
Design AI agents that adapt formality levels, communication styles, humor, and content boundaries for different cultural markets without stereotyping or alienating users.
Implement locale-aware currency formatting, multi-currency conversion, and precise number display in AI agent responses for global user bases.
Learn how to implement the Model protocol in OpenAI Agents SDK to connect any LLM — Anthropic Claude, local Ollama models, or custom endpoints — as your agent's reasoning engine with full tool-calling support.
Learn how to model complex agent workflows as directed acyclic graphs with dependency resolution, parallel execution of independent tasks, and topological sorting for correct execution order.
Implement data loss prevention policies for AI agents that detect and block sensitive data in prompts and responses. Covers DLP policy engines, content scanning with regex and NER, blocking rules, and exception handling workflows.
Build privacy-compliant AI agent systems with data classification pipelines, PII anonymization techniques, retention policies, and consent management to meet GDPR, HIPAA, and other regulatory requirements.
Learn what data processing units (DPUs) do, how they offload networking and security tasks from CPUs and GPUs, and why they are becoming essential in AI data center architectures.
Build a data quality pipeline that validates incoming data, deduplicates records with fuzzy matching, normalizes schemas, and ensures your AI agent's knowledge base stays clean and accurate.
