Hello everyone,
I have been working on an exciting project recently - the implementation of a local agentic AI solution using LlamaCpp. This is a new type of AI model (gguf) that doesn't come with any guardrails, providing a lot of freedom for creative and innovative applications.
I have been working primarily with Python, and I also used cloud agents to help with the development process. So far, I have managed to build a basic framework and have integrated 4 base tools - calculations, file operations, command permissions and execution, and web search. However, there are still a few hiccups. The web scraping function is a bit choppy and needs more refinement.
I also noticed a data feedback loop bug, which makes the agent seem a bit barebones. Despite these issues, I believe that with the right help and collective brainstorming, we can get this agent to work efficiently.
I am currently looking for interested individuals or groups to collaborate on this project. If anyone here is interested in contributing or if you have any ideas on how to improve the agent, I would love to hear from you.
Let's work together to make this agent as functional and useful as possible. If anyone is interested, feel free to reach out to me or post your thoughts here. We can take this project to the next level and beyond.
Beyond this project if anyone is interested in any sorts of collabs lmk!
Thank you for you time!
I have been working on an exciting project recently - the implementation of a local agentic AI solution using LlamaCpp. This is a new type of AI model (gguf) that doesn't come with any guardrails, providing a lot of freedom for creative and innovative applications.
I have been working primarily with Python, and I also used cloud agents to help with the development process. So far, I have managed to build a basic framework and have integrated 4 base tools - calculations, file operations, command permissions and execution, and web search. However, there are still a few hiccups. The web scraping function is a bit choppy and needs more refinement.
I also noticed a data feedback loop bug, which makes the agent seem a bit barebones. Despite these issues, I believe that with the right help and collective brainstorming, we can get this agent to work efficiently.
I am currently looking for interested individuals or groups to collaborate on this project. If anyone here is interested in contributing or if you have any ideas on how to improve the agent, I would love to hear from you.
Let's work together to make this agent as functional and useful as possible. If anyone is interested, feel free to reach out to me or post your thoughts here. We can take this project to the next level and beyond.
Beyond this project if anyone is interested in any sorts of collabs lmk!
Thank you for you time!
