London · Engineer
Umar Salim.
Software engineer, shipping production systems since 2014 across startups, fintech, web3, and AI. I run Salim Consulting, building software and AI systems for clients.
Writing
All 49 posts- 1,844 tokens/s on a 5090. What was actually useful? Qwen3.8 on one RTX 5090: a parallel demo, matched xhigh coding tests, and why I kept my existing production setup.
- Testing Jev as Android Tools for Codex Jev and Gemini Flash Lite responded quickly but slowed Codex's Android workflow. Across 90 runs, neither justified an extra decision layer.
- A First Look at Jev A first look at TypeSafe's Jev, the model that answers questions instead of writing. The cheapest paid LLM on OpenRouter matched it on accuracy for a fifth of the price, but Jev's probabilities were the best calibrated as they came, and twenty questions about one document cost a tenth per answer of asking one.
