When it comes to AI Memory, everyone's arguing "RAG vs. grep" like it's a religious war. It's not. It's just a cost curve. I've gotten this question so many times that I thought I'd just share my thoughts. So here goes:
Vector search wins when your corpus is massive, messy, and unstructured. Thousands of docs, no clean boundaries, meaning matters more than exact words.
Filesystem plus grep wins when your corpus is structured and actually yours. A folder of markdown files you can open, read, and audit line by line. No infra required.
Anthropic already ran this experiment in production. Claude Code dropped its RAG pipeline for plain agentic search (grep, glob, read) and it outperformed the vector pipeline on real work. Not close.
But the benchmark wars are missing the actual point. It was never about picking one. It's about knowing what each layer is for.
Markdown is your source of truth. Portable, human readable, greppable, not locked to one provider. Memory you can actually own and move.
Vector search is an accelerant on top of that truth. A fast index for when the haystack gets too big for exact match to keep up.
Use either one alone and it breaks down:
- Markdown alone stalls at scale and struggles with paraphrasing or fuzzy recall
- Vectors alone turn your memory into a black box you can't read, audit, or export
The next step for memory infrastructure isn't picking a side. It's the filesystem as the ledger and RAG as the index on top of it, so memory stays legible and portable, and still fast when it needs to be.
This is the exact direction we're building with AI Context Flow: markdown as the portable, ownable source of truth, with retrieval layered on top instead of replacing it.
If your memory only exists as embeddings inside someone else's vector DB, that's not memory. That's a lease.
Which team are you on?
Great Idea! Had this issue multiple times over the past week alone. Congratulations on the launch 👏
AI Context Flow
@underdogsteve would love to know what you think when you use it :)
Klariqo AI Voice Assistants
This can save so much time. Having to re-explain the same thing over and over again is tedious af. Love this!
AI Context Flow
@ansh_deb early users are saying it's saving them 5+ hours a week! Very excited how this will evolve when we deploy our MCP servers (soon)
"like it's the first day of school"
I like this line. I wonder if you came up with it originally or used AI! Nonetheless, great concept
@hira_siddiqui1 @justin2025 @yuze_li @mujtabaidrees94 Really like this idea for folks trying to replicate but clearly running into the issues of .. why don't you remember this from our last conversation Mr. GPT.
AI Context Flow
@dzaitzow thank you! would love to know how you feel about the extension after a few days of using it!
I had the pleasure to follow the progress of this powerful tool and I could not recommend more. Strong founders also!
AI Context Flow
@marco_marchesi thank you!
Have been using this product in it's beta phase, it's a super productive tool that can help you get you solutions through AI LLMs in the quickest way possible, prompting is made easy through this.
AI Context Flow
@naveed_siddiqui thank you for being our beta user! We would love to hear how you would want the product to be improved further!
I've been using the AI Context Tool for a while now and it is capital A Awesome - and why is it awesome, because Hira and the team built it to solve a problem that both themselves and their peers were facing. It is simple and, just works. Great product from a really great team, let the unbundling begin.
AI Context Flow
@wheelsfelloff thank you so much Mark! Your mentorship has been instrumental throughout the journey!