Cursor Composer 2 Review: Cheaper Than Opus, Built on Kimi K2.5
Cursor Composer 2 ships at $0.50/M input — roughly 1/10 of Opus 4.6 — and beats Opus on Terminal-Bench. Then a developer found Kimi K2.5 in the model ID.
Posts
Go Iterators: A Practical Guide to the iter Package
How to write a custom Go iterator using iter.Seq and iter.Seq2, handle early termination, convert push to pull with iter.Pull, and avoid the yield traps.
Cyprus Software Engineer Salary 2026: What You'll Actually Earn in Limassol and Nicosia
Cyprus software engineer salary 2026 — real levels.fyi numbers for Limassol vs Nicosia, plus how the new 50% exemption above €55K flips net pay vs Berlin.
Hetzner vs DigitalOcean 2026: Real Numbers After the Price Hike
Hetzner just raised cloud prices 30–37% on April 1. Real migration numbers, current per-month costs, and when DigitalOcean still wins for your stack.
Meta Layoffs May 2026: 8,000 Jobs Cut Starting May 20
Meta is cutting around 8,000 jobs starting May 20, about 10% of its workforce. AI-assisted productivity is the stated reason, with more cuts later in 2026.
Claude Code Subagents: The Practical Guide
Claude Code subagents run side tasks in their own context window. Here's how to create them, when to pick one over a skill, and the mistakes to avoid.
Recursive Language Models: How RLMs Beat Long Context
Recursive language models treat a huge prompt as a Python variable the model can grep and recurse over. MIT's paper shows it beats GPT-5 on long context.
Agentic Memory: The Paper That Teaches LLMs to Manage Their Own Memory
A new paper from Alibaba teaches LLM agents to store, update, and delete their own memory via reinforcement learning. Beats Mem0 and A-Mem on 5 benchmarks.
Polars vs Pandas in 2026: Honest Benchmarks and When to Switch
Polars vs Pandas in 2026: 5-11x speedups on the right workloads, but Pandas still wins for small data and ML pipelines. Real numbers from a 100M-row run.
ty vs mypy vs pyright: Which Python Type Checker Wins in 2026?
ty is 10-60x faster than mypy and pyright but passes only 53% of typing spec tests. Full speed, conformance, and IDE comparison with real project benchmarks.