Blog Archive
Agentic Retrieval: Letting the Agent Decide What to Fetch
Move beyond static RAG with agentic retrieval. Let agents decide what to search, when to stop, and how to combine information from multiple sources.
Golden Trajectories: Curating Human-Approved Agent Runs
Build golden trajectory datasets by curating human-approved agent runs. Use these validated paths to fine-tune agents, create evals, and improve reliability.
The Supervisor Pattern: Coordinating Specialist Agents in LangGraph
Implement the supervisor pattern in LangGraph to coordinate specialist agents. Build a routing system that delegates tasks to expert sub-agents and aggregates results.
Building an MCP Server with the Python SDK
Build a production-ready MCP server using the Python SDK. Expose tools, resources, and prompts to AI agents through the Model Context Protocol.
Prompt Injection Defense for Autonomous Agents: Classifiers and Sanitizers
Protect autonomous agents from prompt injection attacks using input classifiers, output sanitizers, and layered defense strategies for production systems.
Vector Memory for Agents: Embedding and Indexing Conversation History
Implement vector-based memory for AI agents by embedding conversation history, enabling semantic search over past interactions and long-term recall.
Checkpointing Agent State with LangGraph: Threads, Checkpointers, and Resume
Master LangGraph checkpointing to persist agent state, resume interrupted workflows, and build fault-tolerant multi-step agent systems with threads and checkpointers.
Cost Governance for Agent Fleets: Budget Caps, Spend Tracking, and Alerts
A practical guide to implementing cost governance for AI agent fleets, including budget caps, real-time spend tracking, alerting, and cost allocation across teams.
Model Routing for Multi-Agent Systems: Matching Tasks to the Right Model
Learn how to implement intelligent model routing in multi-agent systems using LiteLLM, matching each task to the optimal model for cost, speed, and quality.
Agent Identities in the Enterprise: Service Accounts for Autonomous Tool Calls
How enterprises are implementing identity and access management for AI agents that make autonomous tool calls, with practical patterns using MCP and service accounts.
A Year of Agents: What We Learned Building Production Agent Systems
Hard-won lessons from deploying AI agents in production environments at scale, covering reliability, cost management, evaluation, and the patterns that actually work.
The Agentic Web: Why Sites Are Becoming Agent-Ready
Explore how the Model Context Protocol is transforming websites from static pages into agent-ready interfaces that AI systems can navigate, understand, and act upon.