AVB
@neural_avb
Creator behind Neural Breakdown on YT. Next vids: Multi-agent RL Navigation, Agentic Memory from scratch
Planning a new DSPy video around integrating long term memory into agents. Basically implement the core features of Mem0 from scratch. When you think about it, a good memory system has sooo many moving parts. It’s the ultimate Context Engineering problem. Indexing, filtering,…
Very interesting use case… Without per-user fine tuning, I guess the most obvious way I can think of to reduce latency is to apply continual/lifelong prompt optimization at a per-user level. That way the system prompt has most of the useful user infos at query time. Additional…
the struggle i have run into is latency; working with voice agents, you want as little latency as possible, however, the best memory adds 200-300ms, almost as much as an llm call. that is something that i think will need to be solved properly to have coherent multi turn…
Feeding horses in the middle of a tech demo is such a power move 🙏🏼🙏🏼🙏🏼
I read a lot of open-source code and that’s mostly how I learn to write better code. But it often takes a while to understand different projects their structure, how things are connected and GitHub’s UX isn’t that great for that. built "Gitvizz" to solve that exact problem it…
.@neural_avb explains how to “think the @PyTorch way,” breaking down its role in training and shaping modern deep learning systems. towardsdatascience.com/the-basics-of-…
maybe you dont realize how good you are doing cuz you keep raising the fkn bar.
Never thought I’ll have to try hard to sound human.
It’s so bad. The AI articles are so obvious and typical. These days when I’m writing scripts/articles, I’m actively trying to not sound like AI (writing in a more conversational tone and avoiding the typical AI words and phrases).
There is an article like this every month and every article is slightly different
Clearly we haven’t met
Struggling to grasp how @PyTorch actually trains neural networks? @neural_avb demystifies the core mechanics to help you understand how models truly learn. towardsdatascience.com/the-basics-of-…
For reference, this is what I-PPO would do. Every agent planning decentralized value networks - they won't know what the best action is for the whole group. They still reach and get their rewards, but it's uncoordinated.
United States Trends
- 1. Bears 90.1K posts
- 2. Jake Moody 13.9K posts
- 3. Happy Birthday Charlie 10.7K posts
- 4. Snell 24.8K posts
- 5. Bills 143K posts
- 6. Caleb 49.5K posts
- 7. Falcons 51.7K posts
- 8. phil 181K posts
- 9. Josh Allen 26.9K posts
- 10. Jayden 23.1K posts
- 11. #BearDown 2,395 posts
- 12. Joji 31.5K posts
- 13. #Worlds2025 18.2K posts
- 14. Swift 291K posts
- 15. Ben Johnson 4,466 posts
- 16. Turang 4,373 posts
- 17. #Dodgers 15.5K posts
- 18. Troy Aikman 6,635 posts
- 19. Roki 6,122 posts
- 20. #BeckyEntertainment 113K posts
Something went wrong.
Something went wrong.