AI Agent Frameworks and Platforms in 2026: How to Choose for Production
Compare 20 AI agent frameworks and cloud platforms for production systems. Learn when to choose AgentCore, Google ADK, Foundry, Lang...
Read insightCodeCrux / Field notes
130 insights / and countingAI engineering
in the real world.
Practical thinking on the systems, controls and trade-offs that make AI useful in production.
01Read
what we
are building.
what we
are building.
AIML
Google's Approach to AI Agent Security: Identity, Permissions, and Runtime Boundaries
An introduction to Google's approach to AI agent security: how identity, delegated permissions, and runtime boundaries contain agent risk in production.
Read the latest insight
NIST AI Agent Security: Build a Risk Management and Evidence Plan
Learn how to apply NIST-aligned controls to reduce risk across AI agent identities, tools, and execution paths while building an aud...
Read insight
OWASP AI Agent Security Guide: Translate Agentic Threats into Engineering Controls
Learn how to translate agentic threats into actionable engineering controls with a practical framework for securing AI agent identit...
Read insight
AI Agent Security Governance for Regulated Industries: Policies, Evidence, and Approval
How are regulated industries handling ai agent security? Build policy boundaries, approval gates, and audit evidence for enterprise ...
Read insight
AI Agent Security Audits: A Step-by-Step Assessment Checklist for Production
Learn how to run an AI agent security audit on production agents, testing identity, tool permissions, prompt injection, and fail-clo...
Read insight
AI Agent Security Monitoring: Trace Identity, Policy Decisions, and Tool Outcomes
Learn how to trace agent identity, policy decisions, and tool outcomes in AI agent security monitoring, so every action is attributa...
Read insight
How to Test AI Agent Security: Red Teams, Tool Abuse, and Regression Suites
A hands-on guide to ai agent security testing: map the attack surface, abuse tools in a sandbox, and lock in results with regression...
Read insight
AI Agent Security Multi-Step Tool Attacks: Detect and Stop Chained Abuse
A comprehensive tutorial covering ai agent security multi-step tool attacks: detect and stop chained abuse. Practical examples, best...
Read insight
AI Agent Security Gateways: Enforce Policy Before Every API and MCP Tool Call
A comprehensive tutorial covering ai agent security gateways: enforce policy before every api and mcp tool call. Practical examples,...
Read insight
AI Agent Security Tools: Build a Governed Gateway for Model and Tool Access
A comprehensive tutorial covering ai agent security tools: build a governed gateway for model and tool access. Practical examples, b...
Read insightAI Agent Security Frameworks Compared: OWASP, NIST, and Enterprise Controls
Learn how to apply an ai agent security framework by comparing OWASP, NIST, and enterprise controls to secure identity, tools, memor...
Read insight
AI Agent Security Architecture: Map the Attack Surface and Build Defense in Depth
A comprehensive tutorial covering ai agent security architecture: map the attack surface and build defense in depth. Practical examp...
Read insight
AI Agent Security Best Practices: Identity, Least Privilege, and Human Approval
A comprehensive tutorial covering ai agent security best practices: identity, least privilege, and human approval. Practical example...
Read insight
AI Agent Security Risks: A Threat Model for Autonomous Tool-Using Systems
Model and mitigate AI agent security risks across prompts, tools, memory, and credentials, then apply least-privilege identity, JIT ...
Read insight
What Is AI Agent Security? A Practical Guide to Risks, Controls, and Governance
Learn what AI agent security is, how agent attacks surface across identity, tools, and memory, and how to apply least-privilege cont...
Read insight
Synthetic Data for AI Agents: Generate Training and Evaluation Sets Responsibly
Unlock the power of synthetic data for AI agents by learning how to responsibly generate high-quality training and evaluation sets. ...
Read insight
AI Agent Failure Recovery: Checkpoints, Timeouts, and Durable Execution Patterns
Master robust AI agent design by implementing crucial failure recovery mechanisms. Learn how checkpoints, timeouts, and durable exec...
Read insight
MCP Client Development in Python: Connect AI Models to Secure External Tools
Discover how to develop robust Python clients for a Managed Connectivity Platform (MCP), enabling your AI models to securely interac...
Read insightHave a production AI problem?
Bring us the hard part.
Talk to an AI architect about the workflow, the constraints and the path from prototype to production.