Загрузка видео...

Не удалось загрузить видео

На главную

o1-engineer is here! 🚀 A coding assistant built from the ground up to leverage o1 reasoning capabilities. It can create and edit multiple files or entire folders, plan complex projects, execute them, and write code reviews. All from your terminal 👨‍💻

309,202 просмотров • 1 год назад •via X (Twitter)

Комментарии: 11

Фото профиля Pietro Schirano
Pietro Schirano1 год назад

💬 You can chat regularly or execute commands: /create to create files or folders /edit to edit file, files, or folder content /add to add files or folders to the chat context /planning to create detailed plans /review to create code reviews that you can use directly

Фото профиля Pietro Schirano
Pietro Schirano1 год назад

Repo here. Star the repo so you can keep track of updates and improvements. ⭐️

Фото профиля Pietro Schirano
Pietro Schirano1 год назад

Fun fact about building this. The first functionality I built was /edit, so I used the script itself to keep refining it and understand what sort of instructions work better with o1.

Фото профиля Pietro Schirano
Pietro Schirano1 год назад

So far the biggest performance improvement is in the editing and creation of large files. I was able to edit 2000 lines of Python in one shot with no errors.

Фото профиля Pietro Schirano
Pietro Schirano1 год назад

In order to assure the best results with editing, the script employs an agentic approach, where an instance of o1-mini provides edit instructions, which a second one applies. A paradigm I created with Omni-Engineer, so I brought it back here since it works really well with o1.

Фото профиля Roshan Patel
Roshan Patel1 год назад

can i fork this? applying to yc soon

Фото профиля Pietro Schirano
Pietro Schirano1 год назад

Lmao

Фото профиля nisten - e/acc
nisten - e/acc1 год назад

wait so i can use any openai compatible api to try this on? would .. qwengineer work lol

Фото профиля Pietro Schirano
Pietro Schirano1 год назад

Yes, it's written in OpenAI format, so it should work lol. The only thing is that o1 doesn’t support system prompts, so everything is built as a regular chat completion. And function calling is all made from scratch with json and parsing.

Фото профиля Court Reinland
Court Reinland1 год назад

I like this tool. Thank you. But my initial tests say Claude engineer is still outperforming my simple needs. Claude 3.5 is hard to beat at coding. I tried building same tool in this and in Claude. Claude engineer performed better. Perhaps if I have a more complex need…

Фото профиля Pietro Schirano
Pietro Schirano1 год назад

Yeah I made Claude Engineer too! :)

Похожие видео

OpenAI just announced API access to o1 (advanced reasoning model) yesterday. I'm delighted to announce today a new short course, Reasoning with o1, built with OpenAI, and taught by Colin Jarvis, Head of AI Solutions at OpenAI, to show you how to use this effectively! Unlike previous language models which generate output directly, o1 “thinks before it responds,” and generates many reasoning tokens before returning a more thoughtful and accurate response. It is great at complex reasoning -- including planning for agentic workflows, coding, and domain-specific reasoning in STEM fields like law. But how you should use it is quite different from other LLMs. I think o1 will be a game changer for many AI applications; and in this course, you'll learn how to use it effectively. In detail, you’ll: - Learn to recognize what tasks o1 is suited for, and when to use a smaller model, or combine o1 with a smaller model - Understand the new principles of prompting reasoning models: Be simple and direct; no explicit chain-of-thought required; use structure; show rather than tell - Implement multi-step orchestration in which o1 plans, and hands tasks over to gpt-4o-mini to execute specific steps; this illustrates a design pattern to optimize intelligence (accuracy) and cost - Use o1 for a coding task to build a new application, edit existing code, and test performance by running a coding competition between o1-mini and GPT 4o - Use o1 for image understanding and learn how it performs better with a "hierarchy of reasoning," in which it incurs the latency and cost upfront, preprocessing the image and indexing it with rich details so it can be used for Q&A later - Learn a technique called meta-prompting, in which you use o1 to improve your prompts. Using a customer support evaluation set, you'll iteratively use o1 to modify a prompt to improve performance You'll also learn about how OpenAI used reinforcement learning to produce a model that uses "test-time compute" to improve performance. I think you'll find this course enjoyable and valuable. Please sign up for it here:

Andrew Ng

357,401 просмотров • 1 год назад