Astro Dan

#Python, #AI, #ML, Data Science and Engineering

Let's talk AI for work and productivity.

[Astro Dan @egbo_sky]

EdgeRAG is an agentic #AI RAG for resource-constrained devices. It includes an #LLM, embedding model and vector DB running locally. Check out the project here.

https://t.co/VnHzE5td0H

[Astro Dan @egbo_sky]

@nehulbuilds Just followed you. Follow back.

[Astro Dan @egbo_sky]

AI is reshaping our future. Based on where we are right now with recursive self-improvement, how long would it take before we feel outdated and unable to compete with AI?

What do we do to stay relevant?

[Astro Dan @egbo_sky]

@NateDBusch @austinc3301 @ladymcscope Hi Nate

Thank you for sharing. I am reading all of them. They're very useful resources for people like me.

[Astro Dan @egbo_sky]

What are the best ways to minimise token usage in Opencode and still reach your objectives?

Are there clear guidelines out there one could explore?

[Agus 🔸 @austinc3301]

@egbo_sky @ladymcscope Take the AGI Strategy course by BlueDot! https://t.co/S5lyJCaRW1

And check out https://t.co/3XiAikGt3R

[Astro Dan @egbo_sky]

@austinc3301 @ladymcscope Hi Agus,

Thanks for sharing the link. I already applied for the next cohort of the AGI strategy course.

I am also reading the info shared in the 8000 hours website. It is quite long and resourceful.

[Astro Dan @egbo_sky]

@austinc3301 @ladymcscope Hi Angus

What do you recommend for a beginner to AI safety?

[👩‍💻 Paige Bailey @DynamicWebPaige]

Live demos + hands-on examples of agents in @GoogleAIStudio kicking off tomorrow! 🚀

Join me at @DataScienceDojo's Agentic AI Conference as we take Gemini 4, @NanoBanana 2.1, @GoogleGemma, Gemini TTS and more. If you're yearning to learn more about managed agents, roll through:

[Astro Dan @egbo_sky]

There are so many copywriters talking about AI these days.

I'm not sure I want to be seeing such posts.

[Astro Dan @egbo_sky]

I am still working on this project. Will share further updates soon.

[Astro Dan @egbo_sky]

EdgeRAG is an agentic #AI RAG for resource-constrained devices. It includes an #LLM, embedding model and vector DB running locally. Check out the project here.

https://t.co/VnHzE5td0H

[Astro Dan @egbo_sky]

@adxtyahq You're right man. I'm focusing on my personal project and figuring things out.

[Astro Dan @egbo_sky]

It is difficult to keep up with all the news related to AI.

Hence, why I am looking to connect with people doing great stuff with AI, building agents, testing alignments and evaluating agents/LLMs and working in research labs.

If you're one, let's connect. I follow back.

[Astro Dan @egbo_sky]

What is coming after AI agents?

Comment and follow me.

[Astro Dan @egbo_sky]

@nnennahacks This is beautiful and a motivation to do mine although I'm just a beginner.

[Liquid AI @liquidai]

Today, we announce 𝙻𝚒𝚚𝚞𝚒𝚍 𝚁𝙸𝙿𝙻 (real-time inference pipeline), our ultra low-latency inference engine built to put LFMs directly in the request path of real-time workloads like search, ads, fraud, recommendations, and payments.

> complete generated responses end-to-end in sub 50ms
> maintains low latency under continuous peak load
> LFMs get up to 20x faster on RIPL
> RIPL supports Multimodal LFMs
> RIPL supports d1, our decision models too!

RIPL optimizes the full path from request arrival to a complete generated response. As traffic rises, its performance advantage is greatest on the slowest requests. Let’s see some production use cases:

[Nnenna @nnennahacks]

https://t.co/f5MDBYDU7R

[Astro Dan @egbo_sky]

@akacoding Kindly follow back

[Astro Dan @egbo_sky]

Looking to connect with people doing great stuff with AI. Building agents, testing alignments and evaluating agents/llms.

Say Hi. Let's connect.

[Astro Dan @egbo_sky]

I want to get started with AI alignment, where do I start from?

[Astro Dan @egbo_sky]

@alishaz1221 This is a good habit to practice daily. I think I should start the same.