Blake Rhodes

How Vizient empowers healthcare providers with reliable GenAI insights using LangGraph and LangSmith Vizient's GenAI platform helps users manage data to query for information ranging from patient outcomes to clinical benchmarking. See how they used LangGraph and LangSmith for multi-agent system relia...

So I have been busy at work, but it is so rewarding. Working with the talent at #LangChain to make #healthcare better is the pay off for me.

Long Running Agents | Blake Writes (about) Code From Traditional User Agents to the New Era of LLM-Enabled Agents So, we’ve established that “agents” aren’t exactly new—your browser’s been faithfully acting on your behalf for years, fetching those ...

Hey everyone, I just published a deep dive into the world of long-running AI agents. They’re more than just hype; frameworks like ExACT and CoA are showing us real innovation in how these agents strategize, collaborate, and grow.

blakerhodes.github.io/2024/12/04/l...

It encourages it to not go full steam ahead and think first... And not be so arrogant! 😂

"There was initially 0 squid in the text.
Now there is still 0 squid, but since you asked me to subtract 1 squid (-1), I've actually added 1 to the total count of squid...

So, here's the extra squid: 🐙"

Abliterated models can not refuse the user. So it found away. ollama.com/AutumnAureli...

I am going to try this exercise with an abliterated model.

Agents | Blake Writes (about) Code Marketing Ruins Everything

Agents have been around forever (hello, web browsers 👋), but LLMs are making them smarter. Think: semantic reasoning + tools = better cat pics, fewer existential crises. Just don’t let the buzzword hype distract you from the real potential. 🐾

blakerhodes.github.io/2024/11/27/a...

YouTube video by Melkey Why Cursor AI Just Crushed Github Copilot in the Code Editor Race

Cursor AI Just Crushed Github Copilot in the Code Editor Race
youtu.be/BOxyDT4jJOQ

Continuous Growth | Blake Writes (about) Code Skilling Up in the Age of AI

As the floor is lowered on technology, new talent is empowered and new skills are needed.

qwen2.5-coder The latest series of Code-Specific Qwen models, with significant improvements in code generation, code reasoning, and code fixing.

I think I am in love with ollama.com/library/qwen...
I have been using this as my co-intelligence, and I plan to switch to it. I really like the explanations it gives. This is likely just preference on my part. Anyone else using this mode?