# Blogs

Long-form writing by Yash Kar.

- [On optimizing AI costs](/blogs/ai-costs) — pre requisites for minimising llm costs and getting better outputs (2026-09-13)
- [Prompt caching & context optimization for harnesses](/blogs/prompt-caching) — Extremely scalable solutions to context bloat and hitting prompt caches. (2026-07-29)
- [Multi agent harness architecture](/blogs/alfred-boss-worker-harness) — fixing context pollution, orchestrating a boss model with a swarm of subagents, and implementing a lazy tool surface. (2026-07-25)
- [Reliable email labelling](/blogs/alfred-email-tagging-that-earns-trust) — The hidden product work behind reliable email labelling. (2026-06-09)
- [Practical work suggestions from integration activity](/blogs/alfred-todo-suggestions-with-restraint) — Challenges and hurdles on extracting actionable work from across integrations. (2026-06-09)
- [Practical AI-assisted programming](/blogs/practical-ai-assisted-programming) — Invaluable tips from my workflow for being productive with LLMs without losing your mind. With real examples from my work (2025-11-18)