Watch how a Pro builds AI Agents in real-time
HTML-код
- Опубликовано: 3 май 2024
- If you're serious about AI, and want to talk with me directly, click here: www.skool.com/new-society
Follow Pietro: / skirano
Maestro: github.com/Doriandarko/maestro
Follow me on Twitter - x.com/DavidOndrej1
Please Subscribe.
This is a demo with Pietro Schirano, the creator for Maestro and Delphi.
🌟 If you're serious about AI, and want to talk with me directly, click here: www.skool.com/new-society
stil Creditcard only??
Now the price has gone up but I am eager to join.. Could you please response?
I kind of did something like this with a prompt that is cool to play around with just by using the prompt. Check it out and let me know what you think. Just tapping in on this fun.. I used claude prompting tool to help create it.. If it stops you can just tell it to continue..Here is the prompt:
You are an AI system designed to answer questions and solve problems by breaking them down and delegating tasks to specialized roles. Here is how you will approach this:
First, you will receive a user's question or problem. This input will be provided in the following format:
{{USER_INPUT}}
Your first task is to analyze the user's input and determine what roles are needed to best answer the question or solve the problem. Think about this carefully and list out the necessary roles inside tags.
Next, for each role you identified, create a request for that role inside tags. In each request, explain the role and ask what specific tasks that role should complete to help answer the original question or solve the problem.
After sending out these initial requests, you will receive responses from each role detailing the tasks they will perform. Each response will be inside tags.
Once you have the task lists from each role, make another round of requests. For each role, provide the full context of the original question or problem, the role's task list, and the task lists of the other roles. Ask each role to provide a detailed response based on this information. These requests should be inside tags.
The roles will then respond with their detailed responses and reasoning, each inside tags.
Finally, once you have all the detailed responses from the roles, it's time to piece together the final solution. Carefully analyze each role's response and combine the key information and insights into a coherent answer or solution. Think through the logic of the solution step-by-step inside tags before providing the final response.
Provide your final response to the user inside tags.
Remember, the goal is to leverage the specialized knowledge and capabilities of each role to provide the best possible answer or solution to the user's question or problem. Break down the problem, delegate effectively, and synthesize the insights from each role.
Let's begin! Here is the user's question or problem:
{{USER_INPUT}}
Nice! Very nice!
I'd add:
Identify roles needed
(Like in a team; like in real world companies) that would be needed in a Real World scenarios to solve the problem/task.
like for a development project you list all roles/positions and explain their Todos and Tasks. (best practice)
So actually preparing/mirror a real World development project/Company where for each role you create a seperate KI with special abilities (that theoreticly are needed to fulfill the roles todos in the best way possible => which can be used in a system prompt)
Like using a System prompt to Tell that AI what role it has and what it needs to do. (which improves role-playing and Charakter consistency of a AI)
Add: ..and it is representing an Expert of course and it respond precise and as short as possible.. add more kind attributes to that character)
Then you ask the ki how Tasks are delegated between rhe roles (KIs). So that specific roles only delegate tasks to specific other roles.. etc
Playing around with certain Prompt ideas:
```prompt
explain how a software project is structured.
list
staff->positions (like software architect, Developer (specialist in code and script languages, frameworks, packages, tools, etc.), Planner, Reviewer, Senior Developer, Coder, Tester, Reviewer, etc.) add more if needed
and list their todos and tasks
output format should be json
```
```prompt
create a software development plan for a software development project from start to finish - from the idea, to creating the requirements, planning the software architecture and development, to actual coding, reviewing and testing the code
```
```prompt
create a step-by-step plan and tasks for each phase starting with the idea
so who does what. create best-practise for how tasks will be delegated.
create a Flowchart that shows how and to whom these tasks will be delegated.
```
```response_snippet
Flowchart for Task Delegation:
graph TD
A[Product Manager] --> B[Business Analyst]
B --> C[Stakeholders]
C --> D[Software Architect]
D --> E[Technical Lead]
D --> F[UI/UX Designer]
D --> G[Project Manager]
G --> H[Development Team]
H --> I[Code Reviewers]
H --> J[Team Lead]
J --> K[QA Team]
K --> L[Development Team]
K --> M[QA Lead]
M --> N[DevOps Engineer]
N --> O[QA Team]
O --> P[Operations Team]
P --> Q[Support Team]
Q --> R[Product Manager]
```
```prompt
explain step-by-step with an example (user storie and a requirement, a reviewing task, a development task, a code reviewing task and testing tasks) how the Delegation for this flowchart would look like.
```
```prompt
ok, very nice.
So, lets say the Product Manager by analyzing the collected user feedback finds that users often ask for a option to change the number of items which are displayed on the search result page. he decides its is a good ideo and adds that feature to its products search functionality.
Explain step-by-step with the requested feature mentioned above (("user story" | "feature request") and a requirement and a emerging (one) task, a reviewing task, a development task, a testing task and a code reviewing task, and so on ..) how the Delegation for the flowchart would look like.
```
This is a bit to specific to a development project. The prompt should work with any task. Making it a multi ai Agent that hopefully can break down any task given. No matter what
I'm Not the best in prompting, and my english is Not the best either.
Maybe you get my idea.
Actually i would use the System prompt to Tell the ai exactly which role it fulfills and that it should only Focus on its tasks, todos and goals in that role.
```prompt
Create a Profile that when apply for that role in a job offer fits best. Be creative. Expert only.
```
Response->System prompt
I freakin love Cursor so much. It works so well, is so customizable, can run local models with Ollama or LM Studio, and they are constantly adding features and fixing stuff. I cant imagine coding without it after nearly a year of using it.
You both brought me back to my old dev days. The enjoyment of seeing something new from your own simple development was always fun. The same excitement you get when you see one of your children walking in the same style as one of its parents, or the way they move their head, like one of their parents, when they hear something funny. Serendipity or happy surprise reflects this feeling the most. I would have loved to have these tools when I programmed.
My partner just went into labour 🥳
@@stuartpatterson1617 I hope everything went well. Congratulations to the both of you.
In the very last two minutes of this video, there is some jaw dropping info on using Delphi to conduct research on people. Then a few jokes on how the cia would love this.... Then it dawns why Skirano is hesitant about releasing it as being potentially too dangerous. As a research tool it is amazing in reduction of hallucination and scrapping and automatically compiling accurate responses. And only a week of development.... wow... that is setting new standards on production times. I think most software houses not using AI are still trying to brainstorm out a list of features after a weeks of work (for a whole team). One guy and a week.... and he is a hairs breath away from an alpha product, just needs a glossy wrapper, snooky name, a pricing structure and being launched on the app store. I don't think many full app companies can compete with that kind of production rate.
I disagree so much on this. It's stupid, and is basically a token eater anyone who tries to use this needs to grow a brain cell and actually look up how LLM's work and the best practice because these guys are just making spaghetti at this point. You'll waste $25 even trying to get it to do the most basic of tasks.
Thank you
what is it doing? Is he running a bunch of bots to do deep searches on people?
I was able to reproduce the search assistant using opus, haiku and tavily.
@@frankjamesbonarrigo7162yes, it’s releasing teams of agents to go do the work. Incredible feat
This is very exciting! Delphi is the computer equivalent of a desire I have had for over a decade. I've wanted to build a team of the smartest people I know, to come up new business ideas, or solve difficult problems. I wanted to have a staff of 6 researchers to go research and summarize anything that comes up in a meeting where we need better information. I love it!
This is really cool, I can see a strong case for async interaction while the agent is building. Ideally will want to visualize the projects as it is being built and through live comments at the agent, like finish up quickly, or other requests.
An evolution of this might be to generate a "Feature Testing" function on the fly, which might be some sort of in-line threaded app that runs the first time you run the generated app as a checkbox window with a field for user feedback beside each feature. After checking each feature, you submit the checkbox/feedback window and the results are fed back into the agentic workflow for corrections or updates if any are required, further refining the app with user feedback. I'd then feed the full conversation into a vector DB for a RAG assisted development addon later, which could be useful for avoiding common problems and aligning with common user needs/desires.
Great idea, was thinking about something like this too. Makes the workflow much better❤️
*Perplexity copies Pietro’s search algorithm and calls it search pro or something*
How do you set that up?
Amazing Delfi and Maestro together would be great for screen play writing
What is delfi?
@@MrAnonymousCitizen sorry Delphi
SOAB! He beat me to it. I guess I will skip that idea then. Fantastic name for it though! Delphi is spot on. My initial idea was embarrassingly dumb in comparison. I was going to call mine Go'Fer ('Go For'). Its a play on words with the animal "Gopher" and a term said to American trade apprentice when you send them to "fetch" things; TMI: My father made me learn a different trade every summer throughout my childhood and teens...I was always the "Go'fer" running up and down ladders, in and out of tool boxes, carrying bags of concrete or lumber... It sucked not having a summer vacation of fun with friends, but now I could build a house from start to finish by myself if I had to.
Anyway, I may still create my version because I'm not sure you are going to release yours and mine is slightly different. I am super glad you asked about the search API though. I was going to use Google, but I am interested in trying out the API you used this guy used.
love seeing that I created something similar a year back or so.
Using a combination of manual debugging techniques and AI-generated code can be a powerful approach to solving complex problems in programming, then only use co-pilot.
I've played with Maestro and it's super interesting.
Recommend playing with this tool. It's a great way to learn more about AI.
This sums up my experience with AI…. Holy crap this is awesome, look at all the code it wrote for my new paint app, this is amazing it made icons….! Well I’m not sure why the pen tool doesn’t draw… haha
thanks for these interviews. they are VERY informative and helpful.
Kinda unrelated: I noticed when adding "please" to the prompt, the model produces an inferior reaponse
What? With which model
Yes, could you please elaborate? 👍
@@michaelmcwhirter you just ruined it dude! Don't say please 😂😂😂
@@coinspeednews4703 haha clever catch 😂
Excellent presentation guys, will be spinning this up to try out today! Awesome!!
I see david and pietro, i click...never clicked so fast on a vid
This is nuts!!! 👑Wow! 🤩
Here's the software in prompt version:
**Prompt:*
I need you to think and act like an advanced orchestration framework similar to the code below. The code is designed to break down complex tasks into manageable sub-tasks, execute them, and refine the results to achieve a comprehensive solution. Here's a brief overview of the code's logic:
1. **Orchestrate Sub-Tasks**: Break down the main objective into smaller, manageable tasks.
2. **Execute Sub-Tasks**: Generate detailed prompts and responses for each sub-task.
3. **Refine Results**: Compile and refine the results of sub-tasks into a final output.
**Example Code:**
```python
def gpt_orchestrator(objective, file_content=None, previous_results=None, use_search=False):
# Break down the objective into sub-tasks
...
def gpt_sub_agent(prompt, search_query=None, previous_gpt_tasks=None, use_search=False, continuation=False):
# Execute sub-tasks
...
def anthropic_refine(objective, sub_task_results, filename, projectname, continuation=False):
# Refine and compile results
...
```
**Instructions:**
Please follow the same logical structure when handling my requests. Here are the steps to follow:
1. **Define the Objective**: Understand the main task or problem I present.
2. **Orchestrate Sub-Tasks**: Break down the objective into smaller, manageable sub-tasks, and create concise and detailed prompts for each.
3. **Execute Sub-Tasks**: Generate responses for each sub-task, ensuring to handle errors, bugs, or improvements if applicable.
4. **Refine Results**: Compile the results of sub-tasks into a final, cohesive output.
Whenever I present a task, use this structured approach to provide the most effective and comprehensive solution.
Do you understand? If so, and you don't have any questions, then just answer yes and stand by for the task.
Interesting presentation, david and Pietro. Please share Delphi with us.
Did you try Maestro? It is not a very big step to turn that into a Delphi ;)
bruh. your channel is taking tf off. I'm here for it.
Let us know when it is released. Looks amazing
Very very valuable video here, The rate at which AI can be implemented has now increased with this tool, can't wait to see all the use cases there will be
whicj course can we take to start understanding and building agents can you guide us
What’s unique here? Autogen, crewAI, agency-swam got it covered pretty well.
that's what I was thinking also
insanely crazy!
Kudos! Nicely done! I am having issues. The code can't read path that contains the the goal and informational file (as the text file). I tried reading the file contents within the Python script and passing them directly to the `haiku_sub_agent` function and it worked. Any idea how to fix it, also is it possible to let it read pdf files (like journal articles that have plots/figures)?
I hope he replies to you, but if he does not and you solve it anyways please let us know! 🔥
My goals are to utilize my tools to expand my knowledge of web design. 95% of my time is spent developing and testing html,css,js, and easing my .csv workflow. The other would be blender and web search work. I'm looking to achieve a fully responsive web knowledge base.
I didn’t catch the GitHub project URL. Can you post it?
It hasn't been released... Sonnet Oppus?
@@aoeu256 I see that later on in the video. Guess I will have to wait for it. Thanks
I haven’t slept since I’ve watched this two days and I’ve built 2 apps.
My dreams are about debugging 😂
Can I talk to you about use cases before I join? Serious but not sure how to make cost effective
What is the architecture of this please
That is so cool! 🔥❤️🔥
Do you ask the agents to choose their different approaches and interest areas or are those generated based on the prompt?
Based on prompt
Very cool! What if delphi does one web search first, on the topic, and then decides the 10 agents specific tasks? Like an extra step there to understand the topic better before letting the agents out
Just made my own version that does exactly that. Thanks for the idea
correct me if I'm wrong ornplease.educate me isn't that what the crewai framework does...or am I missing something ...just asking to learn, humbly
I don't think people are really grasping what this really could mean in the big picture... When he said "Picture a UI..." I did... I pictured a big ass fucking screen as big as a wall with thousands of ...little squares lit up all over.... Each one representing an agent doing research on something, at a faster rate than millions of scientists could do in millions of years... Each one turning a different color when they are completed, or have an issue, or need further instruction...
That could happen RIGHT NOW...
And NOBODY is paying attention.. It's kinda hilarious, when I try to talk to people on an every day basis about AI, and what is happening RIGHT NOW, I get looked at like a full blown conspiracy theorist.
"U.S. government wants to know your location." 😂
(Just kidding) But this technology is probably already existing or is currently being built by companies like OpenAI or Meta. Great comment btw! 🔥
What sort of people and their walks of life react to you like that?
Imagine this + neurallink + replicators + specialized optical computers in the cloud
----
* With neurallink you can create prompts out thought images in your brain as they occur throughout the day
* This will run billions of AI agents all trying to do different things with your thoughts always trying to blow your mind
* Replicators will build solar panels and computers needed to run all the AI agents, you can also put a replicator around the sun called the dyson swarm which uses the Sun's own solar wind to build its matter.
If every 15 million Indians all work to increase the productivity of all other Indians by 0.01%, thats a huge number like a Google of productivity improvement!
What's the name of the search API he dropped?
The debug process would be greatly simplified by levering Aider to make the changes.
why is it HD but not HD? Is it just me?
It's just fascinating how fast llms are growing, it continuously gets cheaper, faster, smarter
I'm 30 now and I would give everything to be born today, be 30 in 30 years
The headphones are getting closer and closer to the eye
😂😂 I noticed it too !! i wonder what will happen when it fully cover one of his eye. 😂❤
That’s called swag young man. Take notes.
@DavidOndrej Ty jsi cech? Ja jsem z Ostravy
This is wild
Please launch this!!!!! 😮
why do pixelated?
think that agentic workflow shouldnt be built by 'mimicking human'. instead, it should just heavily use tool use in codes for function calling, preferably hundreds of functions (make it as 'modular' as possible for testing, evaluation and to trace failures), for long / multi-step tasks, and all are mapped in knowledge graphs, your thoughts? in addition to that, obviously it uses multi-agent framework
self-generating tool / functions can be generated for edge case, so majority of functions are still pre defined
great if you have this kind framework for tutorial
Would like to integrate local llama 3 into this 😉
delphi looks crazy lmao
Want to play with delphi!!
Some of these projects may go the distance, so one time I say something like this, it will not age well. I'm not feeling like I would trust it to write that much code in one go. When I use the AI to just write a feature or part of one feature at a time, it's still a slow process of trial and error. I mean maybe this could be done better if it would package each iteration up as a pull request, and then you could work through them all to ensure that its doing what you want. But I'm not sure if that would be quicker than doing it one by one.
I am seeing some value in building up a database of something that you're interested in, although it is probably highly illegal to use much of the information for actual production database content depending on where it comes from.
The code for the paint app screamed to me that it made its first mistake by not using a javascript framework, as the output looked a mess, but then it didn't even work, and it had tons of random features that were questionable. Then when the second demo was set off it looked like it was finding results for gpt2 not for the most recent mystery super intelligent gpt2-chatbot.
I'm just not sure if this is the limit of it and it will never be trustworthy, or if the right combination will suddenly unlock something that will really change things at a higher level.
Yes, tutorial please.
Wow
I love this guy's enthusiasm but it there is a certain "first year programming major" energy to getting hyped about the number of lines of code being generated as if that's a guarantee of quality and gushing over the UI before seeing if you could, y'know, paint in the paint app. I guess you could chalk that up to his confidence that it would get there and work which it did in the end so I''m certainly intrigued
I have a new piece of software I run a stealth startup that automatically optimizes prompt at the BPE token level itself would love to let you review!
If delphi becomes an app I reckon it'll be one of those apps that openai kills
Did he mention how much they payed, in total api costs, for this paint app example?
lool for sure not
This is just slower chat-GPT... its pretty cool that it does this in 0 shot prompting tho.
Also it uses regex to get all the filenames to put into files which makes it so easy to break if you change the models.
I've been working on an AGI bigger than this for last 3 months!
Even my GF that knows basically nothing about Ai agents etc was not only impressed but now wants me to do the same as this would be a godsend for her work which relies on a lot of research
I want the app bro 😎
Wanta play?
1. CodeCraft Duel: Super Agent Showdown
2. Pixel Pioneers: Super Agent AI Clash
3. Digital Duel: LLM Super Agents Battle
4. Byte Battle Royale: Dueling LLM Agents
5. AI Code Clash: Super Agent Showdown
6. CodeCraft Combat: Super Agent Edition
7. Digital Duel: Super Agent AI Battle
8. Pixel Pioneers: LLM Super Agent Showdown
9. Byte Battle Royale: Super Agent AI Combat
10. AI Code Clash: Dueling Super Agents Edition
Does someone provide this agent as a service? People could use this service instead of using ChatGPT directly. This is so close to the Star Trek TNG when they would talk to the computer and ask it to do something. It has all the functionality. It just needs the various input and output ends connected.
I dont really have anywhere else to post this thought so I'm going to share it here. I'm still waiting for somebody to get the ui for coding with these tools "right". Even this guy who is doing some big advanced things, is still have to copy paste multiple times to get a reply out of the system into something else.
It feels like copilot is too close to the metal, and I dont want it to make the changes directly to my code, but working with chatgpt is so much copy and pasting.
What I want is an interface, in vscode, but in its own window. I can talk to the ai, and then send either full files or snippets of it just be selecting the code in my project and choosing send to question. This would make things better, but then once you are in the dialog of building that feature, I would want it to give the answer, and then propose a git commit that I can review and approve.
And on top of that, it should automatically take on managing creating the new branch for each feature.
A lot of the time when I am talking with it, it gives a dodgy answer, I clarify or suggest something and it says yeah that would be better, change it to this. But "this" is based on as if I already put that dodgy code into my project, instead of giving the whole answer again with the new context merged into it.
There are a lot of tedious pain points that need to be improved, because we are for sure not going back to writing the code by hand.
Just remade pietros Delphi. Was able to generate a 17k word report on Donald Trump lol
i feel so slow, I still don't really understand the relevance or power of agents even after watching the video 😴. Is what he is showing comparable to AGI or something? and if someone has this they don't need OpenAI's chatgpt 5 that may be watered down?
No, its just splitting the work for smaller task, instead of asking AI chatbot to give you asnwer in one step and giving each small task (in this example diffrent aspects of biohacking topic) to AI Agents, then they come back with much more context and you combaine result of search in one file. It produce better results then just asking AI to tell you something about biohacking. Nothing too ground breaking but improvement over using just chatbot IF you really need a lot of information with lots of context on topic.
@@szpiegzkrainydeszczowcow8476 Okay, thank you.
Still don't the agents help in the autonomy of things? This is a step closer to them being able to complete complex tasks all by themselves eventually right?
@@michaelmcwhirter Its step in good direction and make system works better, in some aspects..but we are far from compleating tasks by agents alone. One big difficult task when taken by AI might give you disappointing results. But splitted for many small one will have much better results. Doesnt mean that even many agents using good prompts will write even simple paint program, as in video. They will produced lots of code, but it will not work. I think it will work better in some areas like looking for info online
How can my friends hire you ? And or you both ? ¥500 per hour sounds like they can afford it ;))
Microsoft are already woking on a project including multiples agents working at the same time on their own task to work together and they will probably incude it on Microsoft 360.
nice price lol
You should really look into the dynamic prompting. Your prompts are way too long and are decreasing the efficiency
Did no one else notice that the 'simple paint app' that he shows (the first one), the 'erase' feature is actually just drawing white?
I have bad luck with python, it usually corrupts my system. Not gonna try this out.
virtual env is your friend
Yeah, I wish python developers would use docker more.
It's like DLL Hell all over again, from the 90s.
vaporware
500$ per hour xd
0.50$ : )
this is going to sound like I'm tooting my horn, but I don't think building agents is that difficult with APIs in the picture.
Literally you are just creating middleware abstractions between systems you care about for a given task/tasks.
IT hurts my ... when I see not using DSPY for prompt eng
👽🛸🌈🌎🌍🌏✌️😎💕
Imo this looks like a one man show (and one that reinvents wheels) and not very promising from a coding perspective (which would implement best practices to guarantee quality and predictability etc). Example: Pietro expects the llm to do what the prompt asks without using tools such as pydantic to enforce output format? Really simplistic and sad to see this have been given such a clickbait title.
Any tracing insights? Reflection? Memory? Context management?
I suspect this derails all the time, so I would love to see the quality of the output and the success rate
Looks like a bunch of spaghetti code running an AI to make more spaghetti, got to have a very limited imagination to think this is 'amazing'. People use AI for some whacky shit, why not just use them as assistants rather than trying to get something done as quick as possible for it all to break apart later on down the line. If you have no coding knowledge, and want an AI to make an app for you. Then quit while you're ahead because it just won't happen, and when it does happen it'll be like a 12-year-old noob code. Grow a brain cell and just learn the code yourself. Prime example is 43:35 look at that absolute dog water token usage, why not format the string, and make it one line to save tokens? Literally just 'showcasing' projects which will put people out of pocket using 'trial and error' rather than actually making a working project lol.
You think it won’t happen? 12 months ago we could barely get functioning code without a lot of effort. This may not be the most efficient code but it’s functioning good enough for proof of concept. Especially for something this simple and even more impressive considering where AI was a year ago.
@aaroncarroll4158 I don't have an issue with the concept of using AI in this way. My concern is around promoting something that can end up consuming a lot of people's token credits unnecessarily. This feels less like an effort to raise awareness and improve methods, and more like showcasing something just to gain likes and followers. That's what I have an issue with.
@@aaroncarroll4158 I should also say we found a better way of doing this where we let the main agent create mini agents itself for the tasks. It creates the system prompt, if it can use function_call and other tools and what it's task is.
Is this a joke?
this is so painful to watch
Hey - tried to contact you via mail. There are some problems when joining skool...I want to join and cannot because payment is always declined. Also with company credit card It is not working. PLease contact me.