Series

Multi-part write-ups, meant to be read in order.

Building VaultMem

Designing an encrypted memory library for AI agents — client-side encryption, multi-modal and temporal search, pluggable backends, and keeping PII away from the LLM provider.

1. VaultMem — Building Encrypted Memory for AI Agents Mar 21, 2026 · 8 min
2. VaultMem Part 2 — Multi-Modal Memory, Temporal Search, and Pluggable Backends Mar 28, 2026 · 10 min
3. VaultMem Part 3 — Hiding PII from Your LLM Provider Apr 01, 2026 · 8 min

Paper TL;DR

Plain-language summaries of papers I've read, and what each one changed in how I build things.

1. Paper TL;DR: Six Papers That Shaped How I Think About AI Memory Oct 18, 2026 · 6 min
2. Paper TL;DR: Hopfield Networks Is All You Need — Attention Is a Memory Lookup Oct 18, 2026 · 5 min
3. Paper TL;DR: Complementary Learning Systems — Why Agents Need Two Kinds of Memory Oct 18, 2026 · 5 min
4. Paper TL;DR: LoCoMo — How Good Is Your Agent's Memory, Really? Oct 18, 2026 · 4 min
5. Paper TL;DR: Memory as Asset — Whose Memory Is It Anyway? Oct 18, 2026 · 4 min
6. Paper TL;DR: Sentence-BERT — The Quiet Workhorse Behind Semantic Search Oct 18, 2026 · 4 min
7. Paper TL;DR: Hopfield (1982) — Memory as a Valley You Roll Into Oct 18, 2026 · 5 min