Page 41
12 articlesThe Vanishing Blame Problem in AI Incident Post-Mortems
When an AI system degrades, blame diffuses across model, prompt, retrieval, eval, and infrastructure simultaneously. Here is the attribution framework that pins incidents to a specific layer before your post-mortem devolves into 'the model just changed.'
Why Vision Models Ace Benchmarks but Fail on Your Enterprise PDFs
Vision models post impressive benchmark numbers on document understanding, but enterprise teams routinely see silent failures on real PDFs. Here's what breaks and how to build pipelines that survive contact with production documents.
Who Owns AI Quality? The Cross-Functional Vacuum That Breaks Production Systems
AI quality failures rarely stem from bad models. They stem from nobody claiming ownership. Here's how to fix the accountability vacuum before it costs you.
Agent Identity and Delegated Authorization: OAuth Patterns for Agentic Actions
When an AI agent books a calendar event or sends an email on your behalf, it operates under delegated authority. Here's how to design OAuth scope contracts, rotation lifecycle, revocation triggers, and audit trails for production agentic systems.
Agentic Data Pipelines: Offline Enrichment and Classification at Scale
How AI agents change the design of ETL and batch-enrichment workflows — variable compute per record, confidence thresholds as operational contracts, schema design for downstream consumers, and monitoring patterns that distinguish model uncertainty from data ambiguity.
AI-Native API Design: Why REST Breaks When Your Backend Thinks Probabilistically
REST was built for fast, deterministic backends. LLM services are slow, probabilistic, and long-running — and the interface patterns that actually hold up in production look nothing like conventional HTTP API design.
The AI On-Call Playbook: Incident Response When the Bug Is a Bad Prediction
Traditional runbooks break when the symptom is 'outputs feel wrong.' A practical triage decision tree, escalation criteria, and postmortem format built specifically for AI systems in production.
The AI Ops Dashboard Nobody Builds Until It's Too Late
Latency and error rate cover less than 20% of the failure space for LLM-powered features. Here are the five production failure modes your APM dashboard silently ignores — and the signal hierarchy that actually catches them.
Chatbot, Copilot, or Agent: The Taxonomy That Changes Your Architecture
Picking the wrong AI interaction paradigm — chatbot, copilot, or agent — creates architectural debt you can't fix by tuning prompts. A breakdown of the trust models, context-window strategies, and error-recovery requirements that should drive the decision before you write a line of code.
The Cold Start Problem in AI Personalization: Being Useful Before You Have Data
New users have no history, your model has no context, and you're competing against the perception that AI doesn't know them. Here's the engineering playbook for bridging that gap.
Why '92% Accurate' Is Almost Always a Lie
A single accuracy number hides the errors that actually matter. Here's a four-dimension taxonomy — correct, recoverable, harmful, abstained — and a one-page format that gives non-technical stakeholders enough to make the right product, legal, and investment decisions.
The Data Flywheel Is Not Free: Engineering Feedback Loops That Actually Improve Your AI Product
Most teams collect thumbs-up/down and call it a feedback loop. The real infrastructure is implicit signal extraction, weak supervision pipelines, and closed-loop architecture that routes production data back into training without drowning in annotation overhead.
About Tian Pan
I'm Tian Pan, an engineer-founder focused on agentic engineering — building autonomous AI systems and scaling engineering teams. I write practical guides on system design, technical leadership, and shipping with AI agents. Previously an early engineer at Uber, Brex, and IoTeX.
