ModelsBest Model for Frontend Code: Judgement Without a Compiler
Frontend work has no oracle that tells you the output is wrong. What that means for model choice, and how to close the loop with screenshots.
ReadNotes on models, agents, and what inference costs
Practical writing for developers building with large language models — how they work, how to pick one, and how to keep the bill predictable.
Picking between models — and testing them on your own work.
Models
ModelsBest Model for Go: Easy to Write, Hard to Get Idiomatic
Go is the language models write most confidently, and confidence is the problem. Where generated Go compiles cleanly and still races under load.
Read
ModelsBest Model for Large Repositories: Context Is Not Enough
A million-token window does not make a model good at a million-line codebase. What actually determines whether a model can work in a large repository.
ReadModelsBest Model for Legacy Code: Comprehension Beats Cleverness
Legacy work is mostly reading, not writing. Why the model that scores highest on code generation is often the wrong one for a twenty-year-old codebase.
Read
ModelsBest Model for Long Context Tasks: Match the Task Shape
Long context is three different problems wearing one name. Which one you have decides whether window size, attention quality or cache pricing is the real constraint.
Read
ModelsBest Model for Low Latency: Time to First Token Wins
For interactive work, perceived speed is decided by time to first token, not throughput. Which models and settings actually make an interface feel fast.
Read
ModelsBest Model for Migrations: Consistency Over Intelligence
Framework and language migrations are hundreds of near-identical edits. The model property that decides success is consistency across files, not peak reasoning.
ReadModelsBest Model for Mobile Development: The Feedback Loop Problem
Mobile work punishes models differently to backend work. Slow builds, churning platform SDKs and visual output change which model is actually worth running.
ReadModelsBest Model for Pull Request Review: The False Positive Tax
An automated PR reviewer lives or dies on signal-to-noise, not raw capability. What the workflow demands of a model, and how to keep the bot from being muted.
ReadModelsBest Model for Python: Everything Passes Until Runtime
Every model writes decent Python, which is exactly why choosing one is hard. The real differences show up in library recency and runtime failure.
ReadModelsBest Model for Rust: The Borrow Checker Is Your Evaluator
Rust punishes weak models loudly rather than quietly. Why iterations-to-green is the metric that matters and where models predictably fail.
ReadModelsBest Model for Self-Hosting: What Fits on What You Have
Self-hosting turns model selection into a memory problem. Which open-weight models fit on real hardware, and what you give up at each tier.
ReadShowing 13–24 of 119 articles