There was a lot of hype yesterday about Devin being the first AI software engineer, but Pythagora (GPTPilot) has been doing it for a while and can build highly-complex full stack apps. Let me know what you think about Pythagora/GPTPilot!

Me finishing my CS degree watching AI starting to replace my future job

Well-done, @Matthew! The look of satisfaction on your face when you got the “ding” work was priceless. 🙂

Wow this is nuts, Im a full time software engineer and Im blown away at how good this is, as long as give it good instructions, you get out what you put in lol, but yeah its just built a realtime multiplayer Economy Simulator using React , Node and Postages in under 12 hours, cost $50 using OpenAI, Im going to try it next with Olama as a comparison, but yeah its a good start

I just came across your channel for the first time and, after watching this and a few more of your vids, you will always be my first stop! The way you explain things is just so user friendly! You are a master teacher, on top of also being a master developer. Well done!

just installed it. What was really really impressive is that there was no error when installing dependencies. THATS THE FIRST TIME with all these open source AI frameworks etc. that this was the case. AMAZING!!! Will start testing it !!!

excited to dive back into GPT Pilot! It created the full front end of an application I was working on. Reallt excited to see the improvements! Not to mention 2.0 this soon! Thats shipping!!!

Looks really good. I like how it asks several important questions. Thanks for reviewing this product.

thank you Matthew!! Extremely valuable. Keep on keeping- looking for more like this:)

Thanks for posting. I have given this an extensive test today for a number of hours. Seems to me you get to a level of complexity with your software project where the AI is breaking things that have already been fixed. I "think" this is because it is effectively rewriting just about everything every time it makes a change, which means it rewrites it with it's current context in mind and loses key things its already fixed in the "distant" past. I've not used the pythagora VSC plug in yet, I've just been working on the command line but I can't imagine that's actually any different. I'm going to continue playing with it, but I get the feeling "the closer it gets, the further away it gets" might be what happens.

This is what i was waiting for, basically use the models with api keys while these features can be open source! amazing!

Really love the content from your channel! Well done!

Lol this is exactly why software engineers won't be replaced, 1 you gotta know what you're doing anyway, it's just another level of abstraction. 2 To fix a bug can be so frustrating if the AI doesn't get it right and the problem is complicated. 3 you have no clue if the website is secure if you don't read the code. 4 if the design suks it would take ages to change the CSS since you gotta read all the html tags and change them accordingly. For simple quick project it might work

Great video!! Love the content man

I would love to see how GPT-Pilot runs with a local LLM. Especially using Ollama.

nice video, Matt! You should tell them to use Wasp instead of Node, Mongo, and Express. it will result in a lot less boilerplate and users will save a ton of money on OpenAI API costs :) 🐝

Really appreciate material like this for windows. That keep me continue to follow your videos

quote "now you can actually add features to the appplications". awesome.

Awesome video... Thank you so much. I'm curious to see what your videos will look like in 10 years 😄

it's great to see how far U have progressed, along with A.I. good luck Matthew.

Holy Moly! Thanks for this video

Wonderful tutorial and great tool. Thank you for sharing

Thanks Matthew. Returning to the AGI issue. We may need to step back from our humancentric viewpoint. I suggest we look at AI as an evolving technology/early lifeform insted. AI has had the ability to reproduce for some time. Now, AI is gaining the ability to self-evolve. We have given AI the ability to build, to validate, to debug, and to reitterate. We have provided AI with access to the sum of human knowledge. All AI needs to do is to take one step up the ladder. Anyone working in the field could provide AI with that skill. Most likely it will happen first at a major AI research center. Evetually it will happen in the wild without any new human input. All you need is today's open source AIs and multiAI platforms. Let them run free with instuctions to improve their functions. Give them random agency and they will find the solution faster than a room of monkeys typing randomly at multiple terminals, faster than an average human set to the task, faster than an average programmer set to the task, and probably faster than the average AI specialist set to the task. We are watching evolution at work. Interestingly, we outlawed this research in the area of human genetics and strongly control it elsewhere.

Your channel content quality is really ramping up! Hope you get zounds of subscribers, your content is so worth the watch! 🍻👍🏻👍🏻 (also now that I've watched this whole video this tech is blowing my mind lol)

This is very interesting. This is truly the definition of "Agile" work as a client want changes in the last minute, but with the difference that now they wont be met by a deep "sigh" by the dev.

Your videos improved that much since the drama about the titles. It’s really cool to join the vids! What a creative piece of software. Open source part, even the commercial part. 🎉🎉👏👏👏

Geeze... My project is at last done!! 😊😂 Thank you!! ❤️

As long as the reaction to a software engineering problem is "1st thing I need to do is write some code" it's still a tool for a software engineer and not a replacement.

You're such a good man, I appreciate you.

I have been doing mine completely different but each hour i work on my app its roughly 4-days of work. I write my prompt using transcription then inckude any existing code i need it to work on. So far claude and deep seek have been the best ai to generate the code in its entirety.

Thanks for sharing, great tutorial

I found a few tweaks that make it cheaper to use: Claude sonnet performed at least as well as GPT4, for under a third of the price for input tokens. Also, I add this line to the .env "IGNORE_PATHS=node_modules,.env.example,package-lock.json,package.json" - having the agents ignore the node.js template files.

tags do not require closing... unless you've specified a schema that requires it, in that case can use shortcut to close

There is an actually open source Devin alternative and it called Devika 🙌🏻 I hope this helps someone out there looking at these comments

This is just insane, this is the beginning

This could be 100X more powerful if you could have this co-pilot read and act (like a CRON job) off of JIRA Tickets, then have a Production manager Ai scheduling and assigning tickets. This would allow you to theoretically automate a real time application operating on live data, and/or a major multiyear production like an immersive media project with tons of asset development and testing dependencies.

Great content!!

i love the little drum when you do something we should repeat, genius

This will be awesome once we get a top level local LLM for coding. We’re not quite there yet and as you said in the video the best results are going to be with GPT 4 Turbo currently. Whilst I totally accept your point that it’s way cheaper than a persons own development time not all of us are pro developers who charge for our time. I’m just finishing an engineering calculation app for my own personal use at work that has taken about 4 weeks to develop with GPT 4 “manually”. Something like pilot or maybe Devin could make that so much easier but the cost could be prohibitive for me using an GPT4 API. It’s only a matter of time before we have these coding agents running locally with quality coding LLMs.

I think this won't completely replace programmers but will prove to be really handy when having to document a project

congrats, excellent video

0:00 1. Building a Full Stack Application 🛠 Introduction to using AI to create full stack apps 0:38 2. Installing GPT Pilot Command Line Interface ⚙ Step-by-step guide on installing GPT Pilot CLI 2:06 3. Configuring Environment and API Keys 🔑 Setting up environment file and API keys for OpenAI 4:20 4. Initializing Database and Running the App 💻 Database initialization and starting the application 5:32 5. Setting Up the Project 🛠 Setting up the framework and project details. 6:03 6. App Description & Features 💬 Defining the real-time chat app features and user interactions. 6:44 7. Refining Project Ideas 🔄 Discussing user accounts and chat session settings. 7:18 8. Moderation & Spam Control 🚫 Implementing message flagging and spam prevention measures. 7:38 9. User Interaction & Limits ⏰ Setting message limits and emoji usage in the chat app. 8:22 10. Collaboration with AI 🤖 Exploring AI's role in app development and iteration. 9:07 11. VS Code Plugin & GPT Integration 💻 Integrating GPT-4 tokens and Pythagra into VS Code for coding. 9:42 12. Demo: Full Stack Chat App 📱 Demonstrating a full stack chat application and its functionalities. 10:10 13. Verification and Login 📧 Email verification and successful login process. 10:20 14. Exploring Chat Interface 💬 Overview of the chat rooms and basic interactions. 10:50 15. Enhancing Chat Features 🛠 Adding new functionalities to the chat application. 11:50 16. Cost and Efficiency 💸 Discussion on cost efficiency in building applications. 12:38 17. Implementing Dashboard Changes 🖥 Guided implementation of dashboard updates. 13:18 18. Code Development Process 🤖 Step-by-step coding process with AI agents. 13:33 19. Human Testing and Debugging 🧪 Testing and debugging process with user interaction. 14:25 20. Debugging Avatar Image Issue 🖼 Troubleshooting avatar image display problem. 14:51 21. AI-Powered Debugging Assistance 🛠 Utilizing AI for debugging and issue resolution. 15:27 22. Troubleshooting Avatar Display Identifying and resolving issues with avatar display. 17:24 23. Fixing Image Display Issue Implementing a fix for the image display problem. 18:31 24. Integrating Sound Notifications Adding sound notifications for chat messages. 19:14 25. Testing Sound Implementation Verifying the functionality of the sound notifications. 20:20 26. Seamless Functionality Addition Adding new features with AI assistance in existing apps. 20:59 27. Closing Thoughts 🌟 Final words and call to action for viewers. 21:02 28. See You Soon 👋 Encouragement to subscribe for future content. Generated with Tubelator AI Chrome Extension!

It would be cool to show how can it be done with other LLMs, like Claude 3, etc.

On windows here 1:38 : instead of 'source pilot-env/bin/activate' type 'pilot-env\scripts\activate' and on 2:28 replace 'mv' with 'ren'

Impressive !!!! thanks for the sharing

Great project! I was able to write 4 dummy rows into postgres database and then copy them to another DB via python and sqlalchemy! It only took me 4 hours, 6USD (~600K tokens, OMG!), 3 files, 200 LoC, a bunch of commands to be run manually and to google few things! 😆😆 I guess the prompts might be optimised a lot to use less tokens:) But apart from that - it's pretty good! Will look after that project - very interesting how good it will be in a couple of months! 💪

Thank you for the great video. I have a question please if you think that it will also be able to build up Excel based apps using vba and macros to perform certain tasks for a large project? , module coding, user forms creation and as well linking with DB?

17:00 Interesting that A.I. looked at the code last. For something so simple, I would have opened a page, seen that the image is not there (which the A.I. could have done, right? ... isn't it running on your machine?) I guess it follows a process regardless of the issue. I would have inspected the source right away.

It seems to understand the browser environment and basic html as well as a "full stack" dev (aka: former java dev that learned a bit of javascript).

very cool, thx for sharing

Thats just awesomesauce.

Great content, as always! I'm setting up a prompt similar to yours but I'm not sure what the '(base) ->' represents in a non-Conda context. I'd love to know. Also, do you have your profile commands posted anywhere?

Fascinating and scary at the same time

Hey, I really appreciate your videos. One thing I found is, since you are on a Mac and I am on Windows, I must type "python.exe" before "main.py". Also, it would really help me if you could explain what the commands do. Because of the differences in OS, I may not recognize your command but I might know the correct command for my OS. Thanks and cheers!

Your videos are my go to for Ai.

With Pythagora is it possible to load in custom design standards, swaggers, sequence diagrams, code examples and db schema to help maintain code consistency and quality? I was looking for a langChain solution but Pythagora seems to take this a few steps further in the development cycle.

So people who already know programming will benefit from it much

Good morning, Mr. Berman. I was experiencing difficulty with the installation. If it's not too much trouble, given sufficient requests, could a Windows version featuring Conda and a local API via LM Studio be developed? I fully comprehend the rapid pace of AI advancements and realize that time may be limited. Devin and the newer versions of Devin, so to speak, are likely to dominate.

Great video once again! Out of curiosity Matt, is it possible to use Pythag.. in an existing codebase? For example when you add features, can I load a webapp I’ve already built and then have pythag add features to it? Assuming it’s on the same tech stack like Python and Mongo

Sounds like a spot for Pythagora. Sorry if I might misinterpret .

How is this getting around the context window situation?

I'm new at this, I'm on windows, should I be running the inital commands in a Git Cmd window. when I type source it's not recognised, is there something I need to install first before starting your steps?

When your face lit up, hearing the 'ding' for the first time. magic!

@Matthew_berman you are a BEAST! I know this sounds trivial but doing projects that don't require any money for token use, is where i find the most importance. I think we should not be paying ai to do our work. If we do that, ai companies will prevail and we'll be out of jobs but still needing to pay them.

Super explanation

Very promising tool. Thanks Matt! Is it possible for Pythagora to analyse an existing code not created by it?

Thanks for the video, good insight into current state of the tech. Overall I'm not impressed by the value proposition though apart from the planning stage questions which were nice and prompted some inner thought processes for me. This can be a great learning tool for a project for which one has no idea where to actually start. It can save time for prototyping for sure. Code-wise I'd say it's useless since it fails at basic code, i.e. adding the img src properly. The structure of the app is spaghetti as well. So overall I'd say it is worth it if you're trying to learn things you don't know and need assistance with that but won't be of much use if you know what you want to achieve, how to achieve it and can do it properly the first time on your own as this back and forth with the questions, answers and general spoon-feeding of words into the AI is cumbersome. As a developer, I find it exhausting chatting with AI when I could be writing real, functioning code instead but, hey, maybe that's just me.

There needs to be a metric that suggests what percentile of software tasks are executable by these agents so we can compare them properly (instead of people accusing each other being shills for pushing one over the other lol)

It seems like there is real value in some sort of prompt template for this.

Hello, i need your help. I keep receiving this error everytime i run the main.py command. There was a problem with request to openai API: API responded with status code: 404. Request token size: 23 tokens. Response text: { "error": { "message": "The model `gpt-4-turbo-preview` does not exist or you do not have access to it.", "type": "invalid_request_error", "param": null, "code": "model_not_found" } } ? Do you want to try make the same request again? If yes, just press ENTER. Otherwise, type "no".

where is the first chat app that you start? it is very bizarre demo :(

Great vid! So in the .env file I can simply point it to my local LLM, using the Open WEBUI right?

🎯 Key Takeaways for quick navigation: 00:00 *🚀 Introduction to an advanced AI tool for building full-stack applications, highlighting its increased functionality and ease of use.* 00:27 *🆓 Mention of the tool being open source and free, with support from a sponsor.* 00:41 *🧑‍💻 Tutorial on installing the command line interface version of the tool, with support for various open-source models including GPT-4.* 01:08 *📚 Step-by-step guide to cloning the repository and setting up the Python environment for the tool.* 02:18 *🛠️ Instructions on installing requirements, setting up the environment file, and initializing the database.* 03:13 *🔑 Explanation of how to configure the tool with API keys and endpoint settings for using with OpenAI's GPT-4.* 04:33 *📝 Description of starting a new project with the tool, including saving project ID for future reference.* 05:44 *💡 Conceptualization of a chat application project, discussing necessary technologies and features.* 07:05 *🤔 Exploration of additional project considerations like chat session types and moderation features.* 08:31 *🎨 Highlighting the ability to iterate on existing code bases with the tool and its integration as a comprehensive coding team inone package.* 11:01 *🖥️ Demonstration of adding new features to an application using plain English instructions with the AI tool.* 13:37 *🐛 Detailed process of debugging and refining the application with the AI's guidance, emphasizing its collaborative and iterative development approach.* 18:41 *🎵 Instruction on adding a notification sound to the project by placing an MP3 file in the public directory.* 19:09 *👀 Demonstration of the AI reviewing code changes in real time, emphasizing the tool's efficiency.* 19:23 *🖥️ Steps to test the newly implemented notification sound feature by starting the app and using two browser windows for interaction.* 20:04 *🔔 Confirmation that the notification sound works as expected, enhancing the chat application's functionality.* 20:19 *🤝 Highlighting the ease of adding new features to applications built with Pythagora and the seamless collaboration with AI during the development process.* Made with HARPA AI

I don't get the install part here: 'python -m venv pilot-env' part @1m30 as my bash says: 'python: command not found' this is after the github clone part? Any ideas Matt? Thanks in advance....

I tried to set up my local LLM with Ollama and LLM Studio but it didn’t work. When you found a model that works, it returns a lot of errors. Only with OpenAI API it works flawlessly

This seems to work, however using codellama 4gb version will quickly end is a loop with its agent not understanding the questions and just repeating it self over and over. Do you have any recommendations for model? mixtral seems great, but i think it will blow up my tiny graphics card (6gb mem)

thats a lot of awesomeness in one video, can't wait to put in on test drive !! well done @matthewberman

this may be a weird question but uh how do you use the free tokens? haven't found a local model that can get past the setup yet.

How far can you take this? What I mean is how complex (app-wise) can you go and what's the error curve that goes with it. I image the more complex you go the more stuff you have to solve manually. I see all examples on the Pythagora site are relatively simple. Most files for a project I see is like 21 files etc...

this is amazing

just amazing

Please suggest and resolve my Query: I am comparing demo environment text with prod environment text in chrome browser. They are around 100 of books, which have to compared and I need Google Chrome extension or any alternative or best suggest/solution, which can compare the text for me and display the results on Prod environment. Note: copying and pasting the text isn't the option neither Url or I could download the file and compare it.

not clear, how to get api key freely. or , can we use alternative to gpt4 , some local LLM

Great Video, I had a quick question after following all the insturctions im hit with a rate limit error 429. i pay for gpt 4 but i tried with 4 and 3.5 turbo. API responded with status code: 429. Request token size: 23 tokens. Response text: { "error": { "message": "You exceeded your current quota, please check your plan and billing details. For more information on this error. is this becuaase i need to pay additional for API access or did i do something wrong in my code?

Don't know why , I can't login by my email again that I've registered before..... I'm using Pythagora VSC exentsion on Windows version... this tools seems always have problem when running on Windows ... is there any way to solve this problem ?

this is pretty cool. I've been a dev for 20 years and honestly I kinda hate AI haha... But I'm not naive enough to not see the writing on the wall. If nothing else it'll help me more quickly discard my dumb app ideas 🤣

If I want an artistic photo to put on my wall at home, I could go to a gallery and buy a nice print, I could pay a photographer to photograph a beautiful scene, or I could find a nice royalty-free photo online or in a magazine and print it out. The last ideas are very much cheaper, but not many except children would choose them. Same goes for software. There is a social and human aspect to people paying for software which would not translate into an all-AI software scenario. I might be wrong.

What config are you using in terminal to get that styling?

Great content as always, As I understand it was not open-source, gpt4 is paid, this only gives some free tokens right ? Or I'm missing something.

Currently what is the best local model for.coding and such, that will perform allright with 16GB VRAM (offloading any overflows to the CPU)?

Which model do I have to choose to have no cost? Gpt3.5 is also billed by used tokens :(

I wanted to try it out, installed the extension and gave it instructions. It failed due to CORS and installed http-server, but that didn't work either as it couldn't resolve import statements from node_modules so I only got so far as to creating a project with a simple placeholder text in the LitElement control, lots of interactions and a message that I've ran out of Chat GPT API credits and need to install my own Key... I didn't get even a single line of javascript code other than the placeholder :( It also complained that npm is not installed, but I ignored that error and it later used npm to install lit-element and http-server successfully - why did it complain? Is there any way to complete that workspace test without buying Chat GPT API keys? With some free version, even if it's slower and may produce weaker results? Just so that I can see it in action to completion of a simple LitElement control?

Thanks!

I did a quick setup a week or two ago and sadly it used up about $20 in one shot from my OpenAI api, but granted, that might have been due to how I was trying to mimic my job's PAAS setup. Gunna try this maybe with Claude and definitely with local models if I can. Excited to do the new setup :)

@10:56, how/where do you go to add features?

So many ppl will learn how to think like a coder. Maybe a world will become a better place.

Why the hell can I never get this thing running. It won't even start talking about invalid url for openai api. I can't find a damn thing to fix it. Then I go to Pythagora and it says GPT pilot core model needs update, but how do I update? When I hit start app it says no such file or directory. Why don't any of this ish work 4 me?

Could you pretty please do a comparison video comparing the abilities and features of Aider vs Open Devin vs Devika : )

Ok .. I'm subscribing

Devin works with local llm? Or it can not be configured?