Jensen Loke

Research Track

Building a Second Brain

My main research programme: what it takes to build a second brain that works for a person and for an organisation - capture, structure, retrieval, and the knowledge layers that let AI agents act on it. Knowledge graph research is published underneath this track.

Writing

01

Agent memory · Personal infrastructure

The Last Project You Build Alone

Why a second brain should be custom to its owner, independent of any one AI harness, and able to give every future agent durable context.

Read article 02

Knowledge systems - Agent platforms

Why a Knowledge Management Layer Still Matters

May 2026

A practical look at where no KM fails, where connectors and search help, and where a structured KM layer materially improves consistency, reasoning quality, traceability, and permissioning.

Read article

Published research

03

Agent memory · Governance · Qwen

Two Brains, One Operator

September 2026

Why personal and corporate agent memory must split, how each half is governed - server-enforced write fences for the company brain, a permissive universal adapter for the personal one - and how Qwen planned the architecture and audited 1,837 pages of company knowledge on local hardware.

Read paper

Knowledge graph research in progress

Self-contained papers, published as single pages and listed here and in the essays archive.

  • In progress

    What a knowledge graph adds over plain retrieval - ontology design, entity and relationship extraction, and head-to-head comparisons against vector-only search.

Related talk: Human in the Loop Related reading: book reviews