AI Tool Stack
Practical builds, patterns, and lessons for AI systems that survive real workflows: evidence, evaluation, review loops, trust boundaries, deployment, and production architecture.
An evidence-driven agentic research workflow for UK company due diligence and insurance underwriting pre-screening, with source-aware evidence, follow-up research, contradiction detection, and human decision boundaries.
A Document AI workflow for UK motor insurance PDFs, with structured extraction, provenance, conflict resolution, and human review.
Using one model for every AI task turns model choice into a hidden default. Production workflows need explicit model policies for task type, risk, latency, evidence, fallback, and evaluation.
Token waste is architecture debt. Oversized prompts, broad context, verbose outputs, retries, review, and eval runs compound into the cost per trusted completed unit.
Semantic similarity measures intent, not validity. Safe AI caching needs source versions, schemas, model policy, evidence, permissions, review state, and wrong-hit tracking.
When an AI workflow fails, the final answer is not enough. You need the input, route, prompt, model, context, schema errors, evidence gaps, review action, and eval case.
AI failures need designed paths: retry, narrower scope, simpler model, stronger model, human review, partial answer, safe message, or stop.
Launch is not the finish line for an AI system. It is when review queues, failures, costs, user feedback, model changes, and ownership finally become real.
