The State of Software Development in 2026: Trends, Challenges, and Opportunities
Software development is entering an AI-augmented era. This report explores the biggest trends, required skills, and strategic opportunities shaping engineering in 2026.
Thinkingout loud.
Engineering notes, product decisions, and the reasoning behind our work. Written by the people who build it.
Performance
Getting sub-1s LCP on a content-heavy Next.js site
A field guide to the three changes that moved our median LCP from 2.4s to 0.88s — without touching the design.
01Product
The product brief that killed a bad idea in 20 minutes
We shipped a one-page decision framework that stops features from being built for the wrong reasons. Here's the template.
02Engineering
tRPC in production: the things no tutorial tells you
After running tRPC in four different codebases, we've hit every edge case. Here's what actually breaks and how to handle it.
03Growth
Attribution is broken. Here's how we work around it.
Multi-touch attribution models sound scientific. They're mostly fiction. We rebuilt our decision-making around what actually holds up.
04Culture
How we run async standups for a team across four timezones
Synchronous standups were eating 30 minutes a day and producing nothing. Our async system costs five minutes and produces actual context.
05Engineering
Postgres query tuning: five patterns that fixed our p95
We went from 800ms p95 to 95ms on a query serving 40k requests/day. Five patterns, no schema changes, no cache layer added.
06