Product Engineering
This hub joins product discovery with engineering execution: defining the riskiest assumption, instrumenting the user journey, designing recoverable states and shipping changes behind measurable guardrails.
How does the interface help users recover from errors?
Which guardrails make a rollout reversible?
DECISION FRAMEWORK
Define the boundary, then choose the technology.
Start with the decision
Write the action, threshold and evidence required before selecting a metric.
Measure the journey
Track successful outcomes and recovery, not isolated clicks without context.
Ship reversibly
Use staged exposure, guardrails, rollback criteria and a clear owner for cleanup.
CURATED GUIDES
All articles →Continue with practical guides.
AI infrastructure · Anthropic API
Anthropic Models API `line`: stop parsing model IDs
Anthropic added a model-family field to the Models API. Here is how to use it for discovery without confusing family, capability and release policy.
Read the guideAI · Content provenanceOpenAI textGrain: what text watermarking can—and cannot—prove
OpenAI made text watermarking opt-in for select API models. This guide explains textGrain, detection limits and a safe provenance architecture.
Read the guideE-commerce engineering · ShopifyShopify bundles: design the cart transform contract
A production guide to fixed and customized Shopify bundles: composition, pricing, inventory, Cart Transform operations, failure policy and reconciliation.
Read the guideAI agents · Tool evaluationEvaluate MCP tool-using agents: score the whole trajectory
A production evaluation contract for AI agents that discovers, selects, approves and calls MCP tools without hiding unsafe or duplicated effects behind a good answer.
Read the guideAI infrastructure · Anthropic APIClaude prompt caching: stable prefixes that actually hit
A production guide to Claude prompt caching: prefix design, breakpoints, TTLs, invalidation, diagnostics and the metrics that prove a real cache hit.
Read the guideProduct engineering · ReliabilityMeasure reliability at the user journey, not the server
Turn a critical user journey into an SLO, an error budget and a release gate that joins frontend experience with backend correctness.
Read the guideSaudi commerce · Product analyticsZid storefront analytics: build an event contract you can trust
Turn Zid storefront events into dependable product analytics with one versioned schema, idempotent delivery, identity stitching and order reconciliation.
Read the guideProduct design · OnboardingDesign onboarding around the first value moment
Replace forced product tours with a resumable path to one meaningful result, supported by contextual help, saved progress and outcome evidence.
Read the guideSaudi commerce · Catalog architectureSalla product variants: migrate the schema, not rows
Changing a Salla option can reshape every sellable variant. Use stable identity, shadow validation and reconciliation before writing prices or stock.
Read the guideProduct management · DiscoveryPrioritize product opportunities before solutions
A practical discovery workflow for moving from a measurable outcome to user opportunities, risky assumptions and the cheapest evidence that can change a decision.
Read the guideProduct engineering · AI evaluationYour AI evals need a portable product contract
A Product Engineering framework for turning AI evals into portable release evidence tied to user outcomes, cost, latency and reversible rollout decisions.
Read the guideProduct design · Accessible interactionDrag-and-drop needs a second path
A practical Product Design framework for reorder flows that work with pointer, touch, keyboard and assistive technology without turning accessibility into a hidden shortcut.
Read the guide