There is so much speculation around how companies will look after AGI.
One thing is clear, agents will be customizing the user experience and will need a memory that could associate, recall and retrieve at the pace of AGI.
wtf..
launched a research idea last night w my autoresearcher..
woke up this AM to Astra hillclimbing the crap out of it...
claude gave up after naive baseline, and said "its theoretically not possible, lets publish this negative result."
product idea:
claude should tool-call to astra for advice...
Highly disappointed by this post from TechCrunch.
They have chosen to ignore Serval's competitor Atomicwork which has raised $40m+ to date.
Vijay & team don't do the podcast and media circuit like the Serval team.
Terrible reporting wherein they quote a Serval investor who obviously doesn't want the world to know of Atomicwork.
The company has a fantastic set of valley VCs such as Khosla, Battery, Okta Ventures and Indian VCs such as Z47 (my employer), Peak XV, Blume & Neon.
@sabakarimm Timing of a launch always matters. If runway can't last until the market is ready, there's no point in being early. History is full of it and @hiddenlayersec nailed it.
For the last two weeks, I have been playing around with vector databases, embeddings, and memories for agents.
Can see the obvious pros also not so obvious cons.