https://ahmadhsn1.github.io/blog/ Technical write-ups on AI engineering by Ahmad Hassan. https://ahmadhsn1.github.io/blog/vibe-coding-vs-learning-to-code-freshers/ https://ahmadhsn1.github.io/blog/vibe-coding-vs-learning-to-code-freshers/ Tue, 06 Oct 2026 00:00:00 GMT Should you still learn to code in 2026? What vibe coding gets wrong, what AI cannot do, and a 90 day roadmap for freshers in Pakistan and beyond. https://ahmadhsn1.github.io/blog/rag-tenant-isolation-vector-index/ https://ahmadhsn1.github.io/blog/rag-tenant-isolation-vector-index/ Mon, 05 Oct 2026 00:00:00 GMT Filtering by user after a vector search quietly breaks privacy and answer quality. Here is how to filter inside the search, and how to test it. https://ahmadhsn1.github.io/blog/ai-booking-assistant-tool-calling/ https://ahmadhsn1.github.io/blog/ai-booking-assistant-tool-calling/ Sun, 04 Oct 2026 00:00:00 GMT How Aria books, reschedules and cancels through LLM tool calling without owning any booking logic, and how to protect a public endpoint. https://ahmadhsn1.github.io/blog/llm-rate-limits-bullmq-reschedule/ https://ahmadhsn1.github.io/blog/llm-rate-limits-bullmq-reschedule/ Sat, 03 Oct 2026 00:00:00 GMT Retrying a daily quota error five times in thirty seconds only burns attempts. How to tell the two kinds of 429 apart and reschedule jobs in BullMQ. https://ahmadhsn1.github.io/blog/llm-chatbot-pipeline-cost-guardrails/ https://ahmadhsn1.github.io/blog/llm-chatbot-pipeline-cost-guardrails/ Fri, 02 Oct 2026 00:00:00 GMT A public LLM chatbot needs caps, caches, guardrails and fallbacks around the model. The pipeline behind The Copper Larder, and how to build yours. https://ahmadhsn1.github.io/blog/integer-money-row-level-security-postgres/ https://ahmadhsn1.github.io/blog/integer-money-row-level-security-postgres/ Thu, 01 Oct 2026 00:00:00 GMT Why money should be an integer, how one database function makes a sale all or nothing, and how row level security keeps shops apart.