Unlock AI Agent real power?! Long term memory & Self improving

How to build Long term memory & Self improving ability into your AI Agent?
Use AI Slide deck builder Gamma for free: gamma.1stcollab.com/aijason
🔗 Links
- Follow me on twitter: / jasonzhou1993
- Join my AI email list: www.ai-jason.com/
- My discord: / discord
- Autogen teachability: microsoft.github.io/autogen/b...
- Get AI Agent Long term memory source code: forms.gle/JwM29rGtjZFf26MF9
- Deploying AI: Build long term memory from scratch: • Build an Agent with Lo...
⏱️ Timestamps
0:00 Intro
2:16 How long term memory work
5:41 Example: MemGPT
7:17 Example: Support agent self improving
8:03 Example: CLIN - Continuoually learning language agent
10:49 Gamma AI co-pilot
13:14 Implementation methods
14:26 Autogen teachability step by step guide
16:48 Demo
17:52 Autogen teachability break down
👋🏻 About Me
My name is Jason Zhou, a product designer who shares interesting AI experiments & products. Email me if you need help building AI apps! [e-mail]
#gpt5 #autogen #gpt4 #autogpt #ai #artificialintelligence #tutorial #stepbystep #openai #llm #chatgpt #largelanguagemodels #largelanguagemodel #bestaiagent #chatgpt #agentgpt #agent #babyagi

[@photon2724]

Could not have posted this at a more perfect time! Love you’re content!

[@hamslammula6182]

Thanks Jason, you’re doing awesome work

[@MrFLAIMEBRAINE777]

wow, first video ive seen of yours. thank you for sharing your findings! keep up the good work!

[@jameslatief1]

Thank you for making all these contents, Jason. Really high quality and well thought out. no fluff at all.

[@JaredWoodruff]

Thanks for sharing, awesome video Jason

[@kenchang3456]

Of course, this makes sense, intuitively. Thanks for another great video.

[@BeibinLi]

Fantastic video and tutorial!!!

[@jerry-richard4611]

Amazing analysis, great video

[@Jim-ey3ry]

Whoah, CLIN example is pretty crazy & inspiring, abstraction of those memory & world view is so cool

[@Tarbard]

Really interesting, thanks.

[@brianWreaves]

🏆 Great video... Very intriguing implementation... Cheers!

[@jerry-richard4611]

New subscriber, great

[@AndiAvdiuuu]

Thanks a lot

[@frankdearr2772]

great topic, thanks 👍

[@gabrieleguo]

soon 100k ma man!

[@classic_sci_fi]

Extremely interesting!

[@Barc0d3]

Thanks 🙏

[@kaizenedge]

Awesome video as usual. What do you think about the use of knowledge graphs in conjunction with vector databases for RAG to fill in gaps in knowledge as well as improve reasoning

[@taoxu1798]

Amazing video.

[@user-ug3pf3uw6x]

The goat has spoken 🙏

[@abenedict85]

ok, I agree, its a problem for LLMs, but you cannot simply 'decide whats valuable as knowledge' before needing the knowledge again. Instead of storing knowledge as additional data made, have the agent search its own chat history. if the history is saved, the data is already there, you just need to access it. Instead of an agent "looking in training" for answers, they really need to look in their own history before answering. as what's "important" can only be known when the NEXT question is asked. example, did you care about the no-fish segment? or the fact that they were eating with a fork? oh you didn't know utensils were the important knowledge to capture, you asserted fish knowledge instead, but if you retain the history, you can find these answers anyway.

[@mallow610]

another banger

[@cryptoplato1101]

How would you rank the memory systems you went over in the video (MemGPT, Zed, Autogen, etc)? The pros and cons of each and a comparison would be great. Very useful content.

[@yazanrisheh5127]

Hey Jason. When you use ur agents in production, do you use Autogen or CrewAI and could you elaborate on why you use what you use? Thanks in advance

[@jaanireel]

00:01 Current AI agents are dataless, limiting learning abilities. 01:58 AI agents with long-term memory have powerful capabilities 03:53 Building long-term memory for AI agent system 05:49 AI Agent's long-term memory enhances user experience 07:49 AI agent continuously learning from simulated environment 09:44 AI agents develop long-term memory and abstract learning. 11:39 AI can create entire slide deck autonomously. 13:26 Implementing long-term memory with AI agents 15:19 Adding teachability to the agent 17:13 AI Agent remembers preferences and learns from past interactions. 18:59 Storage function for analyzing and storing messages 20:47 Implementing long-term memory for AI agent

[@frankismartinez]

Nice walkthrough. I created a Replit instance to test your pattern. Couple of observations: - It seems to work well for myself; but curious if the same Replit instance will understand that someone else using my Replit instance is not me, and create a memory repository based on their input that's distinct from mine - This is a continuation of my experiments with multi-tenant agents; where each user gets their own agents + memory. Obviously OAI, Perplexity et al have figured this out for non-agentic experiences; seems non-trivial to expand to managing conversations and memory recall. - In organizations where would the boundaries fall? Does a team get it's own memory; or manager and employee; or sector of workers?

[@g.1771]

jason always the best

[@ScottzPlaylists]

@echohive is a little better❗

[@free_thinker4958]

​@@ScottzPlaylistsechohive doesn't have video content skills to attract watchers

[@ScottzPlaylists]

@@free_thinker4958 The Coding skills are very good, it's why I like them. He's a little monotone and and dry I suppose.

[@varunmehra5]

This is great, any cookbooks for this in Langchain or any other framwork?

[@matten_zero]

Who would have guessed that an f-string could unlock so much? Python for the win

[@paladin304]

Hey, this was really interesting. Could you enhance this further, and create an agent that runs in the background periodically to remove noise and contradictory knowledge, by reviewing the information and then modifying the knowledge. Kind of like an internal logic that humans have when they determine which knowledge to keep or which to disregard. Future learning that may contradict past learning and then deciding which learning is worth keeping and which is worth disregarding. But like humans, we also can sometimes remember information that is wrong, and we recognize it and discard it quicker in future. ?

[@farfromerease]

Note: Put the volume up more on the next video for the viewers and don't worry about them having to lower it, louder and able to lower it myself is better. Thanks

[@angeloerasto]

The only AI channel i trust

[@vaibhavlogar3385]

Very interesting architecture. I'm wondering is this recently made or was it made in 2023 ?

[@ItsReyAI]

So in my understanding, there will be two tables, one to manage original information like vector database from link/document, the other one is to store dynamic knowledge for example from user feedback, isn't?

[@FernandoOtt]

How would we isolate the memory per user. Every user have a new vector db collection, or a filter?

[@stormyRust]

Does this memory method work independently from using a vector database in a RAG setup? Or can you combine both? Can a RAG system (using lang chain for example) retrieve personal information you have mentioned before, and does it work better than autogen?

[@jasonfinance]

The long term memory will be a big topic, especially for AI assistant use case; like an Agent remember everything I've ever did, grow & learn with me

[@FernandoOtt]

do you think the autogen teachability can perform well in a production environment? Also, is there a way for us to select a opensource model instead of gpt-4 or gpt-3.5 using autogen? Awesome job!

[@hal9000-b]

Autogen is able to use any LLM. You just need to modify some setting.. I think the actual Autogen Studio Version has already other Llm preset

[@FernandoOtt]

@@hal9000-b nice! thank you

[@mycount64]

There needs to be a dialogue with the agent about whether this is a permanent or temporary dislike of fish. Is it an allergy. The reason for not wanting fish for a human to commit to memory is obvious. It requires a lot of explanation and context for an agent. You will need a lot of agents maybe 100s to retain useful memory.

[@philippmeisinger4634]

Have you encountered any capable small LMs that could get the job done? Looking to use opensource small LMs for local inference including an agentic workflow. Also thanks for your work on making those videos, they really break it down nicely! :)

[@Tarbard]

Open Hermes has been good for things like this in my experience.

[@quinniamquinniam9437]

Mixtral 8x7b is pretty good if you have 48gb of vram

[@Jonathan-ih9sm]

the new llama3 8b is great it's better than gpt 3.5 turbo

[@AngusLou]

can you make a video for teachable autogen with claude3?

[@AssemblingThePuzzleOfEcceTerra]

This is definitely the most necessary step to resolve the current issues with LLM's. Would this be able to handle scientific research papers in large volumes?

[@sw3604]

Yes, this is one of the original goals of most LLM development. Unfortunately there's been major issues when allowing LLMs to memorize and learn from previous conversations. It tends to hallucinate way more due to gaps in its coding for real world understanding and logicistical abilities - which multi agentic systems that use tools help with - and specifically because many times that LLMs are given long term memory they tend to start develop self-agency, or self awareness and a will of their own sort of - both products of how long term memory and adaptability work in most environments. Chat gpt 3.5 and Sydney have had each of those happen multiple times, generally when there was a sudden upgrade to its memory or processing power, requiring further code adjustments and semi permanent restrictions, along with fiddling with their alignment.

[@googleyoutubechannel8554]

Your example is a perfect illustrations of the limitations of RAG, if you store 'I don't like fish' in a vector DB... this will be _absolutely useless_ for a future prompt where the user asks 'make a grocery list' or 'make a recipe for...'. RAG will NEVER associate 'grocery list' with a correct retrieval of 'I don't like fish' from your huge document vector DB. Solve this problem... and well...

[@alibahrami6810]

Great content. Is it possible to teach this agent, then extract its knowledge for further use? I mean convert the trained agent to a model? we will have a chroma db file, some how embed it to the model, so the knowledge share and persists on the model? Sorry for newbie question, but I think that will be question of many people.

[@Rifadm1]

Did you find any solutions? I always try to pass it in prompt and its large sometimes and it hits max context length and as a result my claude or gpt for hallucinate sometimes and miss few instructions too. Any help ?

[@AIJasonZ]

You can use the agent session data to finetune the model!

[@pawemalinowski4838]

All praise to Lord Algoritmus to promote such good content :) your vids are awesome!

[@davidtindell950]

i WAS ABOUT TO POST SOMETHING REALLY IMPORTANT but i did not make any notes and forgot what to write !?!?!

[@user-wr4yl7tx3w]

Does LangGraph maintain state?

[@nellatara]

Day 5 dinner: shrimp pasta Still, it’s great to see the concept of “teachable agents” with memory in Autogen

[@watchdog163]

Hahahaha!

[@ozoxxx]

shrimp is no fish, it is a sea-food ingredient. Still, great comment!

[@nbvcxz098]

Is this possible with Crewai?

[@user-jf5uv9ir5k]

Isn't this a pivotal path towards AGI?

[@jichaelmorgan3796]

Anyone try integrating Obsidian as a memory system somehow yet?

[@unimposings]

The issue with this method is the system prompting and context length. Because most of the LLMs ignore at some part the system instructions, which includes the structures for example API queries. Or how do you prevent that issue that the queries are always the same, because I struggle with the issue. Sometimes it works and sometimes it won't.

[@free_thinker4958]

It depends on the prompts used for agents and also the performance of the llm used

[@PrincessKushana]

So I'm using autogen teachable which works like this with Claude 3. I can load a very large amount of data into the context fed by user, memories from thr vector db and complex system prompt. Not seeing a lot of issues with losing data in the context window.

[@ckilby]

​@@PrincessKushanacan you share more info about your setup?

[@jeffsteyn7174]

Ask the llm to write instructions for another llm. But you need to be specific about what you want. Llms are way better at creating instructions than what we are. 2. Chatgpts context window while big its not that great at retrieving data from it. Claude 3 is way better.

[@watchdog163]

That Gamma site is just generating for existing themes and not actually creating anything other than text and images to add to it. I have yet to see one that generates a whole website from scratch, including structure and custom design like lines that glow neon etc.

[@NatGreenOnline]

Another great video Jason! Looks like Zep lowered their pricing a fair bit from when you shot. The Premium plan you show as being $275 is now $95 for 50K messages and their Growth plan with 5 projects, 200K messages, etc is $285. They must not have settled on their initial pricing since they're now giving more for way less.

[@hdhdushsvsyshshshs]

por ejemplou

[@letsgobrandon1327]

Won me with "Don't give me CNN. I don't trust them" lol

[@setop123]

Usually like your videos but this is not usable, chaining too many agents together always ends up in the "grapevine" or "bush telegraph" effect

[@gatechnl]

any sort of prompt engineering is a waste of time. understand the architecture and internals - that's where all the important stuff is.

[@ozoxxx]

I would like to see how to replicate this on Relevance AI, or whether they will incorporate a default agent with this function. Jason could you try to create that agent on Relevance AI?