- Видео 195
- Просмотров 359 254
Automata Learning Lab
Добавлен 1 фев 2021
What up folks! My name is Lucas, I am an AI Engineer and in this channel I want to explore topics like AI, automations and how to build some cool tools!
Automating Standalone Python Scripts with UV & Claude Projects
In this video, we dive into the fascinating world of building Python tools using cloud projects with the custom instruction feature. Inspired by the genius Simon Willison, we explore how to cook up a one-shot prompt that can whip up a standalone Python script for us. We first unravel the mystery around cloud projects and the UV packaging tool, then set about creating a project with the custom instruction to generate Python scripts with inline metadata for UV. We put this to the test by developing a script that merges audio and transcribes from video files, and another that creates a grid visualization of images, all done seamlessly using a standardized approach. Join us on this coding adv...
Просмотров: 108
Видео
How to Use Google's NotebookLM for Studying and Research
Просмотров 1,9 тыс.21 час назад
In this video, we take a dive into Google's Notebook LM, with it you can upload pdfs, Google Drive files and much more to create a centralized research hub. It is powered by Google's Gemini 2.0 to generate concise summaries, insightful briefings, and even engaging podcast conversations about your sources. We'll explore the interface, featuring unique chat and studio windows, and showcase some o...
Getting Started with Gemini 2.0
Просмотров 31214 дней назад
In this video, we dive into Google’s Gemini 2.0 Flash Experimental, a cutting-edge multimodal AI tool available in AI Studio. Be sure to like, subscribe, and stay tuned for more AI adventures! Cheers!. 📚 Chapters: 00:00 - Introduction to Gemini 2.0 00:14 - Overview of Gemini's Features 00:30 - Exploring the Talk to Gemini Feature 01:57 - Demonstrating the Show Gemini Feature with Objects 04:17 ...
Writing Detailed Research Reports Using Gemini 1.5 Pro with Deep Research
Просмотров 75021 день назад
In this video, we dive into the powerful capabilities of Gemini 1.5 Pro with deep research to create remarkably detailed research reports. Don't forget to like and subscribe. 📚 Chapters: 0:00 - Introduction to Gemini 1.5 Pro Research Capabilities 0:10 - Overview of the New Research Report Generation Feature 0:30 - First Example of a Generated Research Report 1:00 - Inspecting the Report and Maj...
Building Custom LLM Researcher Agents with Telegram, LlamaIndex & Python
Просмотров 43221 день назад
In this video, we're going to build a simple automation system that integrates OpenAI's capabilities with Telegram for managing tasks, leveraging llama-index for creating useful agents. We start by creating a custom distributed pipeline capable of assigning different tools like web searching, reading, and writing files to custom agents created during runtime. We guide you step-by-step on settin...
Getting Started with AIDER: AI Code Generation for Terminal Projects
Просмотров 523Месяц назад
In this video we'll go over the basics for getting started with aider-chat for building simple apps directly from the terminal. Perfect for anyone looking to blend command-line prowess with a little AI magic! Don't forget to like, subscribe, and keep an eye out for more coding adventures! Cheers!. 📚 Chapters: 00:00 - Introduction to Aider Chat and Overview 01:10 - Setting Up the Python Environm...
New AI Tools | Visual PDFs + Analysis Tool with Claude 3.5 Sonnet & ChatGPT Search
Просмотров 651Месяц назад
In this video, we dive into the intriguing world of advanced AI features, spotlighting the latest updates from both ChatGPT and Claude (CLA). Including: Claude's new data analysis tool, which lets you create eye-catching, interactive dashboards with ease. We'll walk you through setting up dashboards, tweaking visualizations, and making them more effective. Get a taste of the new LATEX rendering...
Streamline Your 3D Printing Workflow: Automating with Cursor AI, Python and ChatGPT
Просмотров 169Месяц назад
IN this video we'll work on a simple automation for a 3D printing queue using Python, leveraging the Cursor AI editor and ChatGPT. Like and subscribe if you like the video! :) 📚 Chapters: 00:00 - Introduction to 3D Printing Queue Automation 00:50 - Defining the Automation Goals 01:46 - Planning the Print Queue Command 03:22 - Designing the Queue Update Logic 04:28 - Defining Queue Management Fe...
Organizing Your AI Prompts with Jinja Templates with ChatGPT & OpenAI
Просмотров 6042 месяца назад
In this video we'll look at prompt management with Jinja templates and learn how to streamline your workflow with reusable prompt modules. We'll guide you through setting up basic call scripts for chat models like GPT, and then show you how to transform static prompts into dynamic templates that save you time and hassle. Explore the power of systematic prompt reuse and customization, whether it...
Building a Local Voice AI Assistant with Llama 3.2 & OpenAI Whisper Turbo 3
Просмотров 4,1 тыс.2 месяца назад
Building a Local Voice AI Assistant with Llama 3.2 & OpenAI Whisper Turbo 3
Getting Started with ChatGPT Canvas & The New OpenAI Realtime API
Просмотров 1,8 тыс.3 месяца назад
Getting Started with ChatGPT Canvas & The New OpenAI Realtime API
How to Interact with PDFs Using the New ChatGPT 01-preview Model from OpenAI
Просмотров 1,8 тыс.3 месяца назад
How to Interact with PDFs Using the New ChatGPT 01-preview Model from OpenAI
How to Use AI to Write Code | Tips & Techniques to Use ChatGPT o1 Claude 3.5 Sonnet & Cursor
Просмотров 1 тыс.3 месяца назад
How to Use AI to Write Code | Tips & Techniques to Use ChatGPT o1 Claude 3.5 Sonnet & Cursor
Using the New ChatGPT O1 Model from OpenAI to Improve my Python Code
Просмотров 1,5 тыс.3 месяца назад
Using the New ChatGPT O1 Model from OpenAI to Improve my Python Code
ChatGPT O1 Model Released by OpenAI | First Look
Просмотров 4193 месяца назад
ChatGPT O1 Model Released by OpenAI | First Look
Automating Research Tasks with AI | LLM-Augmented Research Workflows
Просмотров 6453 месяца назад
Automating Research Tasks with AI | LLM-Augmented Research Workflows
Building The Same App with Claude + Artifacts & GPT Engineer
Просмотров 6254 месяца назад
Building The Same App with Claude Artifacts & GPT Engineer
Automating High Quality Research Reports with LLMs | From Prompts to LangGraph
Просмотров 7114 месяца назад
Automating High Quality Research Reports with LLMs | From Prompts to LangGraph
A Quick Introduction to LangGraph for Building Agentic Workflows Using GPT-4o-Mini
Просмотров 1,4 тыс.5 месяцев назад
A Quick Introduction to LangGraph for Building Agentic Workflows Using GPT-4o-Mini
Building ChatGPT's App Ideas with Claude 3.5 Sonnet & Artifacts
Просмотров 1,7 тыс.5 месяцев назад
Building ChatGPT's App Ideas with Claude 3.5 Sonnet & Artifacts
Quick Introduction to Fast Tool Calling with Groq and Llama3
Просмотров 4965 месяцев назад
Quick Introduction to Fast Tool Calling with Groq and Llama3
How to Use Claude 3.5 Sonnet & Artifacts for Studying, Learning & Research
Просмотров 9 тыс.5 месяцев назад
How to Use Claude 3.5 Sonnet & Artifacts for Studying, Learning & Research
Building Interactive Apps with Claude 3.5 Sonnet & Artifacts
Просмотров 15 тыс.6 месяцев назад
Building Interactive Apps with Claude 3.5 Sonnet & Artifacts
From Text To Visual Diagrams with LangChain, Pydantic & ChatGPT
Просмотров 6436 месяцев назад
From Text To Visual Diagrams with LangChain, Pydantic & ChatGPT
Automating High Quality Quizzes from Video with Whisper & GPT-4o
Просмотров 3066 месяцев назад
Automating High Quality Quizzes from Video with Whisper & GPT-4o
Automating The Setup of a Github Repo with ChatGPT, LangChain & Pydantic
Просмотров 7057 месяцев назад
Automating The Setup of a Github Repo with ChatGPT, LangChain & Pydantic
Understanding Scientific Papers With Prompt Engineering & The ChatGPT API
Просмотров 5997 месяцев назад
Understanding Scientific Papers With Prompt Engineering & The ChatGPT API
Prompt Engineering for Code Generation
Просмотров 1,2 тыс.7 месяцев назад
Prompt Engineering for Code Generation
Structured Outputs with Pydantic & OpenAI Function Calling
Просмотров 1,7 тыс.7 месяцев назад
Structured Outputs with Pydantic & OpenAI Function Calling
A Simple Prompt Engineering Experiment Template for Text Summarization
Просмотров 7357 месяцев назад
A Simple Prompt Engineering Experiment Template for Text Summarization
very nice. thank you. (feedback: the colorful shape around the pointer is distracting)
@@NardiPaffon ah ok fair point! Thanks ! :)
Greetings! I have been looking for an Artificial Intelligence tool that is feasible to study programming. It seems interesting this NoteBook LM tool. Although it does not matter if I have to pay although of course here they say it is free. But really what I am looking for is that it works and can help me.
Nice ! Then you should check it out and see if you like it :)
Interesting, but not ready for prime time. Firstly, what you show here does not resemble what one sees on Google's site. Next problem is in trying to get the data out. Google seems to randomly disable the copy function. Then there is the faux podcast fearure. What reseacher wants to hear two immiganary entities talk to each other in a contest to see who can sound the dumbest? Finally, it is Google, so we can expect it to be dropped without notice when they get bored.
Right I get it, I think since it is free it’s a nice tool to try but perhaps not rely on it given the points you raised
@automatalearninglab I think the biggest plus is that you can limit the scope that is used for the analysis. Asking Copilot, for example, a question can be infuriating. I asked for a mailing address, and when it came back malformed, the second try was completely different. So was the third. The whole concept of crowd sourcing is flawed. My 1911 Britannica has articles by humanities top thinkers. Now we have Wikipedia spreading nonsense people heard on TV. Not the step forward I'd hoped for.
Thanks for sharing. One little problem I have with NotebookLM is the formatting of the output. It's really hard to read. You have to copy and paste it into another problem like Docs or Office just to be able to read it.
Yah I guess it’s meant to be like a temporary vessel where you develop thoughts and insights then transfer to another medium. Gemini 1.5 pro research has this cool feature of exporting a research report directly to google docs
One thing i’ve been questioning myself it’s if we can delimitate the podcast subject. I have a lot of sources and i would like a more narrow podcast on a specific topic, not with every font together as a whole. Thank you for the video
the only way i found for this is to select less sources in notebooklm or take a look at the llama implementation of this podcast feature which requires some coding experience but gives you in theory full control
Thank you for the recommendation and the great explanation L.
@@user-ox1ud8zn9g you’re welcome! :)
Amazing article and your black tank top
@@marcanttins yessss
Very good!! Tks
@@ElisabeteGeargi hehe
Show!!
@@maggabarbosa thanknyouuu
Please provide code Link of code in discription is not open
Restated almost line from line from the paper for many central claims, hypotheses, and experimental details. Explanations that deviated from wording in the paper were extremely unclear and borderline incorrect.
Yes part of the point was to give a sense about the paper, however, could you point me to my unclear or incorrect comments so I can take a look and perhaps even write a comment here readdressing? Thanks for the feedback :)
Bro, your videos are impressive Love from iran 🇮🇷
Thanks man! That's really nice! :)
"Just like we discussed" Not everyone landed here because they watched your other videos.
@@nas8318 true, that was a slip on my part! Thanks for the tip
what is that eye tracking thing, do you use that for controlling the cursor?
@@alibalbars812 na it’s just a nice app called cursor pro :)
Cê é tão lindo 😍🥰🥰
@@beatrizbelbut4862 hehe ;)
I lost you at "if you want a setup like mine" then my terminal seemed usless and my path all wrong. Actually at 5:03 when you quit the terminal and continue elsewhere.
ok
Had to resort to watching this because ChatGPT is outdated on its own systems, I updated openai package to the latest version and got an error that is never seen on the internet, so they are just making changes on the fly on what and what not to use: "You tried to pass a BaseModel class to chat.completions.create(); You must use beta.chat.completions.parse() instead" oh really? this is the first time on the internet history somebody has got that error and talked about it? and at the time of posting (7 months after upload) response_model is not even a thing, their docs point to response_format and passing the pydantic model there and I get this damn error, which is my fault still, but it doesn't make sense that you change the structure of your implementation every at least 7 months.
Thank you , you think it can’t hold on to longer conversations like meetings and summarize them ?
@@chintrupal4512 it can but perhaps you’re gonna have to optimize the model behind it, I’ll try to do a follow up to this soon
Thanks looking forward to it
source code does not open((((
ill check that soon! sorry about that
Congratulations !
yessss
Very good!!! Tks
eu preciso saber \e.e/ você é Português ou Brasileiro? não tem sotaque quase nenhum kk mas o Tchu(two) e o Dabliu(W) me colocaram essa pulga atrás da orelha e agora eu tenho que saber . Otimo video , estou começando no estudo de IA's pra criar um assistente pro meu homelab
@@ChargeGame top mano! Sou brasileiro! :)
@@automatalearninglab qual a extensão de autocompletar o codigo ali que você usa ?
@@ChargeGame Nao e extensao, eu uso Cursor (IDE top que e um fork do VSCode)
please share code through the github.
Very good. Congratulations !
:)
Great video! Tks
thankssss
Great explanation, RAG architecture would be a video by itself indeed
@@user-ox1ud8zn9g yah!
Crazy!
@@user-ox1ud8zn9g haha it’s pretty nice
Great content as always, if you write a book, guides, or articles please add the link and I will go right away
@@user-ox1ud8zn9g oh thanks! I will! I did a course about prompt engineering if you want to check that out and support: automatalearninglab.thinkific.com/courses/prompt-engineering-basics
👍🏼 Thanks for the video. Well done. I love going from simple concepts and then expand into more generic abstractions because if you understand the underlying logic it's easier to develop and debug later on (A.K.A. having good foundations to build on).
You're welcome! Yeah this video ended up being a real success and I just did it in a whim without thinking too much about it. Thanks a lot! :)
@@automatalearninglab For me, and I guess for others too, it's the incremental explanation that does it. I was looking for a way to understand how tools work. And your video gave me that and also two other, more high level, views on how to do it (OpenAI code and Langchain). I would say this approach is what made the video interesting and engaging...
Nice! Thanks for that lovely feedback! :)
This work is so well-crafted, congratulations!
Very Nice!
Very nice!
Very Nice.
Very nice!
This work is fantastic!
Very Nice !
Truly impressive work!
Your work is extraordinary!
You did a very creative job, it's incredible!
I'm impressed, it’s amazing!
You did an excellent job - it’s amazing!
@@ElisabeteGeargi ooh thank you
Great job! Congratulations!
@@ElisabeteGeargi thanks!!
Great job! Thanks.
You did an excellent job! This is so inspiring.
This video is amazing! Great work!
This video is amazing! Great work!
Very Good!!!!! Thanks !
Very Good !!!!! Thanks !!
Very Good! Thanks!