Varun Nair

Co-founder | Co-founder | Previous: Head of audio software @ Meta, co-founder of @twobig_ears (acquired by Meta) + more.

[Weavable AI @weavable_ai]

Hello World! We are on Product Hunt.

AI agents are only as good as the work context they're given. Right now there's no layer that tracks cause and effect across your work data over time.

Weavable is that layer. One MCP endpoint, continuously updated.

https://t.co/MUc9joYaBW

[Varun Nair @ntkeep]

@BarakDesigns Thanks Barak!

[Varun Nair @ntkeep]

Just launched Weavable on Product Hunt.
AI agents are only as good as the work context they're given. Right now there's no layer that tracks cause and effect across your work data over time.
Weavable is that layer. One MCP endpoint, continuously updated. https://t.co/ZBtoGQOghR

[Varun Nair @ntkeep]

@levie One part of that equation: https://t.co/ZBtoGQNIsj

[Varun Nair @ntkeep]

Better data architecture beats bigger context windows every time.

[Varun Nair @ntkeep]

When using foundation models for product features, the bottleneck shifted from "how do I train a model?" to "how do I prepare data that actually works?"

[Varun Nair @ntkeep]

LLMs care about how you structure information, not just what you feed them.
What you exclude matters more than what you include.

[Varun Nair @ntkeep]

@boztank Boz, it’s amazing to see all the things that were just ideas a few years ago in various meetings become reality. Well done to all (and all those reorgs were clearly worth it 😁)

[Varun Nair @ntkeep]

@ID_AA_Carmack Ah the good ol’ days 😃

[Varun Nair @ntkeep]

@gdibner Gil, not sure if it’s just me but I can’t see any of the links in your recent posts. I only see the text summary.

[Varun Nair @ntkeep]

@vedikaja_in This is great. A few adds:
* Pack everything into the prompt but GIGO still applies. The more relevant the data, the better the output.
* Set up a product push loop. It builds habits fast.
* Different parts of the product should aim for different LLM quality/latency tradeoffs.

[Varun Nair @ntkeep]

@thomasforth It’s actually originally from Inverness and was named “Shropshire Blue” to sound more popular

https://t.co/r7GMNSQ58v

[Varun Nair @ntkeep]

@gdibner Definitely. Although similar to writing, there’s a different kind of thinking that happens when thinking and typing simultaneously. I sometimes rewrite generated code because it helps me get to a result that’s even better.

[Varun Nair @ntkeep]

A startup really is about all your good decisions making the bad decisions look worse

[Varun Nair @ntkeep]

Link here https://t.co/cvnZdUnN2o

[Varun Nair @ntkeep]

I wrote about how I’m using our own product to move faster and to quickly distill what I need to focus on.

[Varun Nair @ntkeep]

@thomasforth Aah versioned releases make me nostalgic. (Kudos on a massive update!)

[Varun Nair @ntkeep]

More here https://t.co/6JAROItUwc

[Varun Nair @ntkeep]

Context is a bottleneck when working with LLMs, not prompt engineering.

[Varun Nair @ntkeep]

@thomasforth Fail.