What an LRS Does and Why an LMS Is Not Enough
An LMS manages assignment, launch, and completion rules. An LRS manages the learning record contract. Here is why that split matters in real xAPI systems.
Discover more
Production Failure Modes in AI Agents and How to Anticipate Them
AI agents usually do not fail with a dramatic crash. They fail quietly through wrong tool calls, invalid arguments, retry storms, looping behavior, and weak recovery. This article explains where the agent loop breaks and how to design traces, guardrails, limits, checkpointing, idempotency, and evals that catch incidents before users do.
Shared State Is Incorrect State: Reduce Hidden Dependencies
The most difficult bugs often come from code that changes data you did not realize it could touch.
From SCORM Runtime to xAPI Statements: The Mental Model Shift
xAPI is not SCORM with JSON. The real shift is from browser-owned runtime state to event records sent to an LRS, which changes launch, identity, debugging, and reporting.
Responses API, Agents SDK, and the Modern Agent Stack
Most teams think they are choosing a framework. In practice, they are choosing what they will be able to see, control, and recover when an agent fails. This article explains the execution layer, orchestration layer, state tradeoffs, and the real architectural decisions behind a modern agent stack.
Breaking Temporal Coupling: Design Beyond Sequence
Systems become fragile when correctness depends on invisible timing assumptions.
Is SCORM 2004 Still Relevant in 2026?
SCORM 2004 is not the future of e-learning standards, but it is still part of the present. If you work with enterprise LMS platforms, legacy content, or structured learning paths, SCORM 2004 remains highly relevant. Not because it is modern, but because it is still deeply embedded in the ecosystem.
How to Build a Useful Eval Set for a Tool-Using Agent
Learn how to design an eval set for a tool-using agent using trace-level evaluation, dataset splits, layered scoring, and realistic failure cases that catch regressions before production.
Configuration: Separate Policy from Mechanism
If changing a setting requires changing code, your system is tighter than it needs to be.
Debugging SCORM 2004 Like a Developer
SCORM 2004 does not fail silently. It tells you exactly what went wrong. If you are not reading error codes, validating API calls, and tracing lifecycle state, you are not debugging. You are guessing.