Erik
Bylund.
A self pro-claimed ambivert with insatiable curiosity over people, things, and systems. Driven by empathy, motivated by technology, inspired by craftsmanship. By trade, a Staff Solutions Architect at Apollo GraphQL, solving complex API orchestration challenges for many of the world's largest enterprises. By practice, trying — slowly, often in public — to help us meet the AI transition without losing what makes us human.
What I am thinking about, in long form.
- No. 01·July 8, 2026
Agents Need Sleep
Distillation concentrates error as efficiently as signal. The memory that makes your agent useful is rotting, and no search index will save it. Part 3: consolidation as a first-class engineering practice.
9 min read - No. 02·July 8, 2026
When AGENTS.md Isn't Enough
The industry calls it a memory harness. I think memory is the wrong word. Recall was never the hard part. Part 2 of a series on agent-ready projects: the salience problem.
11 min read - No. 03·July 8, 2026
Making a Codebase Agent-Ready
Every new agent session is a brilliant coworker arriving with a clean slate. Most codebases are hostile to that coworker. Part 1 of a series on agent-ready projects, memory, and why agents need sleep.
11 min read - No. 04·May 13, 2026
Anyone Can Be Anything (Except Accountable)
Your brilliant, amnesiac coworker doesn't hold the bag — you do.
20 min read - No. 05·May 5, 2026
AI Fluency: A Map for Outsiders and Native Speakers
A framework for locating yourself in your relationship with AI — Outsider, Tourist, Conversationalist, Fluent, Native Speaker — five recognizable places, not a ladder to climb.
28 min read