INSIGHTS

Field notes on AI, software & modern teams

Practical writing on shipping AI systems, engineering craft, and the choices that shape modern software teams. From our team to yours.

AI & Agents

Evaluating LLM agents: three metrics we actually use

Beyond eyeballing outputs — the small, boring metrics that catch regressions before your users do.

Aug 12, 2026 · 8 min read
AI & Agents

The retrieval quality gap: chunking mistakes we keep seeing

Most RAG problems are retrieval problems in disguise — how to spot them and what to fix first.

Jul 28, 2026 · 6 min read
Product

Design systems for AI product surfaces

Streaming, citations, refusals, and repair — the new primitives every AI product interface needs.

Jul 10, 2026 · 7 min read
Cloud & DevOps

GPU inference costs: the tuning knobs that matter

Batching, quantization, and routing — where the real savings hide once you get past the sticker price.

Jun 24, 2026 · 9 min read
Product

Streaming UI patterns for LLM apps

How to make partial output feel intentional, not glitchy — the small state machines behind a good streaming UI.

May 30, 2026 · 6 min read
Engineering

Contract-first APIs: why we default to OpenAPI

The one habit that keeps front-end, back-end, and mobile teams from stepping on each other for a whole quarter.

May 08, 2026 · 5 min read
AI & Agents

Fine-tuning vs. RAG: a decision framework

A short set of questions that tells you which one your problem actually needs — and when you need both.

Apr 14, 2026 · 7 min read
Cloud & DevOps

SRE for AI: what changes when your model is the SLA

Latency budgets, drift detection, and rollback plans — an operations checklist tuned for LLM-backed services.

Mar 02, 2026 · 8 min read

Want more? Get the field notes.

One or two pieces per month. No spam.

Subscribe to Insights →
Building with AI? Zikosoft ships production-grade agentic systems with governance built in. Talk to our AI team →