---
title: "What I'm NOT doing"
url: https://memory.wiki/q13ymHpq
updated: 2026-05-14T18:15:49.480Z
hub: https://memory.wiki/hub/demo
bundle_count: 1
concept_count: 12
source: "memory.wiki"
---
# What I'm NOT doing

> Negative space matters more than the positive list when the work is wide. Pinning this somewhere I'll re-read.

## What I'm not building

- **A vector store.** We use pgvector because Postgres was already in the stack. We are not in the business of competing with Pinecone / Weaviate / Chroma. If pgvector ever becomes the bottleneck, we'll swap. We won't try to be the vector DB.
- **A knowledge graph DB.** Same logic. concept_index + concept_relations live in Postgres tables. We never sell anyone on "a graph database for AI memory" — that's not the product.
- **A fine-tuned model.** Every model we use is off-the-shelf (Anthropic Haiku, OpenAI text-embedding-3-small, GPT-4o-mini for capture, Claude Sonnet for analysis). Custom training is not on the roadmap for v6 or v7.
- **An AI provider.** Obvious but worth stating. We are not building inference infra.
- **A workspace.** No team channels, no rich-text collaboration, no comments threads. Hub-shaped, URL-shaped, single-author by default.
- **A comments system.** A lot of users will ask. The answer is no — comments turn a memory layer into a social tool, and the two have different incentive shapes.

## What I'm not chasing

- **The Notion shape.** Blocks, databases, embedded widgets, real-time collab. Notion does that better than we ever could; we're not in that lane.
- **The Roam shape.** Block-level references, bi-directional links as first-class, daily notes as the entry point. The audience is too narrow and the model is too far from URL-as-memory.
- **The Obsidian shape.** Local file vault, plugin ecosystem, vault-as-product. We're URL-shaped; the Obsidian shape requires we run on disk.

## What I'm not optimising

- **Conversion rate.** Not at this stage. The product has to be the story before the funnel matters.
- **MAU.** The wrong vanity metric for a memory tool — single-user retention over weeks is more meaningful than concurrent monthly users.
- **Viral coefficient.** I care that the dogfood loop works; I don't care about K-factor math.

## What I'm not delaying

- **Security review.** Even at this stage, a bug in shared-doc permissions is a launch-killer. Every PR touching auth or sharing gets a security pass.
- **Founder dogfood.** I have to live in mdfy.app full time. If I drift to using Notion for my own notes, the product gets worse.
- **The Show HN draft.** It gets rewritten every two weeks until launch.


---

## Concepts in this document
- **mdfy** _(entity)_
  A tool that stores project context and decision history, integrated into Cursor via custom rules.
- **Structural moat** _(concept)_
  The defensible advantage created by mdfy's vendor-neutral positioning, which competitors cannot replicate due to inherent conflicts of interest.
- **Claude** _(entity)_
  Anthropic's AI model cited as an example of vendor lock-in through projects and memory features.
- **Hacker News** _(entity)_
  Primary GTM channel chosen for high ICP overlap with developers and format alignment with technical credibility.
- **Cursor** _(entity)_
  Code editor that consumes mdfy bundles as context for chat and composer sessions.
- **ChatGPT** _(entity)_
  Example of an AI provider whose memory feature is intentionally confined to its own product.
- **Cross-AI memory** _(concept)_
  The core thesis: a memory layer that persists across multiple AI providers simultaneously, structurally unavailable to single-vendor AI companies.
- **Anthropic** _(entity)_
  Incumbent AI vendor mentioned as competitive threat with integrated memory and user lock-in.
- **Show HN Launch** _(concept)_
  Central launch strategy focused on Hacker News as primary channel.
- **Launch Strategy** _(tag)_
  The overarching theme covering all aspects of the product launch preparation and execution.
- **mdfy product** _(entity)_
  The core product being developed with build-in-public approach and AI integration as substrate.
- **Competitive Positioning** _(tag)_
  How the product differentiates from alternatives and establishes unique market position.

## Concept relations (within this doc's concepts)
- **Launch Strategy** encompasses approach **Show HN Launch**
- **Competitive Positioning** includes strategy **Structural moat**
- **Structural moat** enables differentiation **Competitive Positioning**
- **Hacker News** primary channel **Launch Strategy**
- **Cross-AI memory** creates advantage **Structural moat**
- **Claude** developed by **Anthropic**
- **mdfy** integrates with **Claude**
- **mdfy** solves problem for **ChatGPT**
- **mdfy** integrates with **Cursor**
- **Structural moat** enables **Cross-AI memory**
- **Cursor** positioned to deliver **Cross-AI memory**
- **mdfy** integrated into **Cursor**

## Bundles containing this document
- [Launch strategy: Show HN week](https://memory.wiki/b/DlrrKcQ2)
  > Pre-launch plan + GTM channels + what not to do. The strategy bundle a reviewer can read end-to-end.

_Hub canonical:_ https://memory.wiki/hub/demo
_Concept digest:_ https://memory.wiki/raw/hub/demo?digest=1&compact=1