Your AI Wrote the Code. Who Understands It Now?
Generated code keeps the artefact and throws away the argument. What to record while the reason is still in someone's head, so the next change is safe.
3 min read
Read
aidrop.it blog
Notes on keeping software worth changing.
Product updates, implementation details, and field notes on getting an app running, who may reach it, and the record the next change starts from.
Latest
From the team
Articles, shipping notes, and the thinking behind how aidrop.it works.
ProductOps for AI Teams: Keep the Product From Becoming a Pile of Prompts
ProductOps gives AI teams one operating system for customer signals, product decisions, releases, evaluation cases, and ownership.
5 min read
Read
You Built an AI App. Now Where Does It Run?
AI makes it easy to build a working app. This guide covers the deployment decisions that turn a localhost demo into a service people can use.
5 min read
Read
AI Agent Skills: The Quality System Behind Repeatable Work
Skills give AI agents a repeatable way to work: the right context, the right sequence, evidence checks, and clear points for human judgment.
5 min read
Read
Case Study: How a Consultant Delegates First Drafts to a Memory-Backed Agent
Case study: a consultant delegates first drafts to an agent built on her recorded standards — it drafts in her format, she keeps judgment. Drafts start at 70%.
3 min read
Read
Your Second Brain Should Talk to Your AI
Your AI can't read your Obsidian or Notion notes. Distill the durable decisions out of them into a memory your agent can query across every tool you use.
3 min read
Read
Git for Your AI's Memory: Why Agent Knowledge Needs Versioning
Facts change, so AI memory needs versioning — history, audit, and rollback, not one overwritten value. Why agent knowledge should work like git, not a notepad.
3 min read
Read
Build vs. Buy: Choosing an AI Agent Context Layer
Choosing an AI memory layer: build on a vector DB, adopt a framework (Mem0, Zep, Letta), or connect a hosted MCP memory. Decide on curation and portability.
3 min read
Read
How Multiple AI Agents Share One Context Layer
Multiple AI agents coordinate best through one shared, persistent memory they all read and write — not lossy message-passing. How shared memory scales a swarm.
3 min read
Read