Home / Articles
ArticlesNotes on product, AI and engineering.
Short, practical write-ups from our work — what we've learned building AI features and shipping products.
AI
When not to use AI: a five-minute feasibility check
A simple way to tell whether a problem genuinely needs AI — before anyone writes a line of code or spends a budget.
AIFrom prototype to production: making LLM features reliable
Evaluation, guardrails and monitoring — the unglamorous 80% that decides whether an AI feature survives contact with real users.
EngineeringChoosing a database for a real-time analytics product
How we weighed read patterns, scale and cost when picking the data layer for a client's customer-facing analytics module.
EngineeringEvent-driven architecture: lessons from a logistics platform
What worked, what we'd change, and when message queues are genuinely worth the added complexity in a client's product.