5 tiers of long-term memory and personalization for LLM applications (in-person workshop)

  Рет қаралды 4,781

LLMs for Devs

LLMs for Devs

Күн бұрын

Пікірлер: 21
@manfredmichael_3ia097
@manfredmichael_3ia097 Ай бұрын
I think there should be a microphone in the middle of the audience. You had an insightful discussion with them, amazing audience!
@devlearnllm
@devlearnllm Ай бұрын
Great idea!
@Bragheto
@Bragheto 4 ай бұрын
This is so valuable! The idea of embedding enrichment with summarizations will be very useful to one of my projects. Thanks a lot for sharing! Congrats from Brazil!
@devlearnllm
@devlearnllm 4 ай бұрын
Glad you found it useful
@lovehappykk
@lovehappykk 4 ай бұрын
‌‌Wonderful narration
@ajax_davis
@ajax_davis 3 ай бұрын
Maybe S tier would be using Reranking (Jina) to also select memories based off the users intent.
@IdPreferNot1
@IdPreferNot1 4 ай бұрын
Thanks for sharing these meetings!
@devlearnllm
@devlearnllm 4 ай бұрын
Hope you found it useful
@danieldesenna7611
@danieldesenna7611 2 ай бұрын
Great video!! Thanks for sharing the code! One question though: Inside A:tier code -> "print_ai_answer" function, you wrote: for like in extracted_personality["likes"]: text_to_embed = f"The user likes {like}" current_embeddings = embedding_client.embed_query(text_to_embed) dislike_with_metadata = { "id": str(uuid.uuid4()), "values": current_embeddings, "metadata": {"type": "likes", "content": like} } embeddings.append(dislike_with_metadata) Was it not supposed to be something like "likes_with_metadata ={...}" and then "embeddings.append(likes_with_metadata)" ? I guess repeating "dislike_with_metadata" does not make a difference for the code functionality, but it was a bit confusing to understand the code for a moment. Thanks!
@devlearnllm
@devlearnllm 2 ай бұрын
Good catch!
@Priming-AI
@Priming-AI 4 ай бұрын
Adorei as aulas, foi muito claro e direto ao ponto, Obrigado
@microburn
@microburn 4 ай бұрын
The camera swivel is an L when you get excited / nervous
@devlearnllm
@devlearnllm 4 ай бұрын
Yeah it’s a bit too sensitive
@__________________________6910
@__________________________6910 24 күн бұрын
No, I know why you want long-term memory-because your AI girlfriend doesn't remember your past conversations. She forgets you, that's why.
@devlearnllm
@devlearnllm 11 күн бұрын
T.T
@lifeoftomi_
@lifeoftomi_ 4 ай бұрын
Great stuff man. What’s S-tier though? Maybe I missed it
@devlearnllm
@devlearnllm 4 ай бұрын
I forgot to mentioned the S tier at the end. It's
@uddhav.m
@uddhav.m 3 ай бұрын
Can anyone share the anthropic system prompt different, can't find it anywhere
@rexsybimatrimawahyu3292
@rexsybimatrimawahyu3292 4 ай бұрын
Is the camera automatic tracking to ur face? Im new into machine learning?
@devlearnllm
@devlearnllm 4 ай бұрын
Yeah, it's the DJI Pocket 3
Attention in transformers, visually explained | Chapter 6, Deep Learning
26:10
How Strong is Tin Foil? 💪
00:26
Preston
Рет қаралды 84 МЛН
💩Поу и Поулина ☠️МОЧАТ 😖Хмурых Тварей?!
00:34
Ной Анимация
Рет қаралды 1,6 МЛН
Worst flight ever
00:55
Adam W
Рет қаралды 15 МЛН
Magic or …? 😱 reveal video on profile 🫢
00:14
Andrey Grechka
Рет қаралды 90 МЛН
LangGraph 101: it's better than LangChain
32:26
James Briggs
Рет қаралды 73 М.
Scrape any website with OpenAI Functions & LangChain
24:10
LLMs for Devs
Рет қаралды 46 М.
No, Einstein Didn’t Solve the Biggest Problem in Physics
8:04
Sabine Hossenfelder
Рет қаралды 300 М.
SF Growth Engineering Meetup - Tools & Frameworks 4/21/2015
52:16
Why Agent Frameworks Will Fail (and what to use instead)
19:21
Dave Ebbelaar
Рет қаралды 61 М.
How to Do 90% of What Plugins Do (With Just Vim)
1:14:03
thoughtbot
Рет қаралды 895 М.
ICML 2024 Tutorial: Physics of Language Models
1:53:43
Zeyuan Allen-Zhu
Рет қаралды 21 М.
Let's build the GPT Tokenizer
2:13:35
Andrej Karpathy
Рет қаралды 605 М.
How Strong is Tin Foil? 💪
00:26
Preston
Рет қаралды 84 МЛН