Content
@
https://warpcast.com/~/channel/firstdraft
0 reply
0 recast
0 reaction
Jason
@jachian
Agents are reflecting and learning Feels just like yesterday that OpenAI launched o1 to crickets.. only to be out-headlined by DeepSeek and their new reasoning model. I think it's easy for us to dismiss what reasoning models are capable of.. "just prompt it a little different". We still compare them against their language model brethren. They're different Dig into the training method and you realize company after company hit ceilings on their ability to throw parameters and more data to infinitely get their models.. smarter.. I've said it before and I'll say it again. I'm less impressed the by the models and their ability to sound vaguely human, but really impressed by their ability to understand us and what we say. Now they can see and hear. But now.. they're reflecting (?) Next frontier that we're just getting a glimpse into is having our agents be able to think ahead and learn from past mistakes. We saw the primitives of it with Yohei Nakajima trying to push the boundaries of the Pippin framework..
4 replies
2 recasts
13 reactions
Jason
@jachian
But now.. leading up to this weekend I met a woman at the AI conference who's a little soft spoken. Another AI founder with a small AI company with an open source framework. She thanks me for giving thoughtful input why inferior frameworks like CrewAI (sorry not sorry) are gaining traction by blitzing the field with marketing. I learn she's a professor at Columbia who teaches about NLP I go further down the rabbit hole and I discover her virtual talk on the workshop day at the conference: How to Improve Your Agents. Casually tops the leaderboard for VisualWebArena for evaluating multimodal agents. They've discovered that not only can you equip your agents to reflect, but to explore their surroundings and learn from their mistakes. We're just seeing the tip of reflection in this moment We are so early
1 reply
1 recast
6 reactions
Nico.cast🐱
@n
Everything is agent
1 reply
0 recast
2 reactions
Brown
@crimsonking
Have you tried o3 mini-high
1 reply
0 recast
1 reaction
Garrett
@garrett
Agentic economy is just now beginning in earnest
1 reply
0 recast
1 reaction
