Writing on LLM memory, RAG & Darija NLP
Notes on LLM memory, RAG, Darija NLP, and building products in Morocco.
- Sep 12, 2026Modeling People as First-Class Entities in Agent MemoryFlat memory can't answer relational questions like 'how has my relationship with X been trending?'. The entity-layer design behind iRemember: alias resolution across languages, evolving profiles, sentiment timelines, and privacy by construction.2 min read
- Sep 5, 2026Compressing Conversation History Without Losing What MattersNaive summarization destroys exactly the details users ask about later. The compression strategy behind iRemember: episode-vs-fact split, progressive oldest-first gating, temporal anchor preservation, and a probe-question evaluation method.2 min read
- Aug 29, 2026Inside a LangGraph Memory Pipeline: From Message to MemoryA teardown of the extraction pipeline behind iRemember: why LangGraph, how a chat message becomes structured memory candidates, deduplication, async processing, and the schema decisions that make retrieval work.2 min read
- Aug 22, 2026LLM Agent Memory: The Complete Engineering GuideEverything required to build production memory for LLM agents: the four memory operations, hybrid retrieval scoring, write control, history compression, person-entity modeling, and how 2026 benchmarks actually compare.5 min read
- Jul 12, 2026Memory Is the Missing Layer of LLM ApplicationsContext windows keep growing, but they still forget. Why persistent, structured memory — not longer contexts — is what makes LLM assistants actually useful.1 min read
- May 3, 2026What Building a Multitenant ERP Taught Me About SaaS ArchitectureTenant isolation sounds like a database decision. It's actually an application-wide discipline. Lessons from shipping a multitenant ERP.1 min read
- Feb 18, 2026Notes on NLP for Moroccan DarijaDarija breaks every assumption standard NLP pipelines make: no standardized orthography, code-switching mid-sentence, and almost no labeled data. Field notes from building sentiment and transcription tools.1 min read
