Featured · Engineeringcache
May 23, 2026·2 min
How KV Caching Dramatically Cuts Your AI Costs
How KV caching works in LLMs, why it matters for AI-assisted coding, and how Nilux keeps about 96% of a request's input in the cache.
Read the post →/blog
Field notes from a terminal-first lab.
Engineering posts, release notes, retros from production. Written by the people who built the thing \u2014 never by marketing.
6 postsRSS · /feed.xml
Researchagents
October 28, 2026·11 minAI coding agents compared: how they differ and how to choose
Nine AI coding agents side by side: where each runs, which models it uses, how you pay, and how it remembers your project, from each maker's own docs.
Company$/tok
May 12, 2026·4 minPay-Per-Use vs Subscription AI Coding Agents: How to Compare Costs
Compare AI coding agent pricing with a practical workload test. Account for included usage, token rates, cache reads, overages, and completed tasks.
Productnpm
May 10, 2026·5 minGetting Started with Nilux AI: From Install to First Task
A step-by-step guide to installing Nilux AI, connecting to your first session, and completing your first coding task in under 5 minutes.
Research2026
May 8, 2026·8 minAI Coding Assistants in 2026: A Developer’s Guide
A comprehensive look at the AI coding assistant landscape — from IDE plugins to CLI tools — and how to choose the right one for your workflow.
Companyv0.1
May 1, 2026·6 minWhy We Built Nilux AI — And Why It Matters for Developers
The story behind Nilux AI — why we built a terminal-native coding assistant with honest pay-per-use pricing and project memory that persists across sessions.