NEW EPISODE
AI Is Moving at 100 MPH. Your Company Probably Isn’t. ft. Sreedhar Peddineni
S4E22
GTM troubles? Tap into this one!!
https://t.co/oRY0kka39b
Spencer's Dev Corner
------------------------
AI powered dev becomes increasingly fascinating and powerful by the day. Lately I am very interested in rapid verification loops that run as close to the code edit as possible. A core principle I am orienting around is "Make AI code fail as fast as possible." An F1 car needs brakes to handle corners at high speeds, agents need feedback at superhuman speed to deliver on true autonomous coding promises.
we got a good one for you guys!
Logan Kelly, CEO of Waxell AI, jumped in to talk agent observability and reliability.
TAP IN
YT: https://t.co/dJWI7IzSBD
Spotify: https://t.co/rWkRZMWgwF
Apple: https://t.co/3FkZvEUMsm
Stop trusting LLMs just because they're useful. 🛑
Our Head of DevRel @RoieSchwabco breaks down the two essential parts of a reliable AI stack:
⚙️ Reasoning: What the LLM does (logic/synthesis/formatting).
📚 Knowledge: What the Vector DB provides (traceable/authoritative facts).
If you can't trace a claim back to the source, you aren't building a knowledgeable app -- you're just vibe-coding.
"Knowledge is the thing that needs to be traceable, explainable, and authoritative."
Ground your model's reasoning in a verifiable data layer. 🛠️
Full @ai_rebels episode: https://t.co/N2DwgEt1Z8
Mythos model card is out, and so is our new episode! Barbara Wittman joined the show to talk about the human layer in AI adoption, and how to focus your AI strategy as an organization.
As always, links are below!!! Tap in 😤
The conversation continues! AI Rebels linked up with Bhaskar Sunkara of @BicycleAI_ to discuss the coming explosion of always-on agents, and how to build the organizational trust structure that enables them to be effective contributors to your team
LINKS BELOW!!!
"All the LLM ever does is hallucinate. It just so happens that a lot of the time, the things it says are reasonable." - Roie from @pinecone, quoting Andrej Karpathy on our latest episode! We dug into RAG, agents, AI memory, and why trust is remains an unsolved problem in AI.
Links below!!
Most of our guests have said the same thing:
You can’t outspend OpenAI or Anthropic on general purpose AI.
The play is specialization.
Pick a specific problem. Solve it better than the foundation models can. That’s the moat.
You don’t need to be better at everything. You need to be undeniable at one thing.