Jefferson Nunn: Good morning y'all. been working on quite a few projects over the last couple weeks. ⁓ Not had a chance to really record and let you guys know what's up. There's a lot of exciting developments, a lot of exciting products that are on the horizon. And one of the things I wanted to share with y'all is actually how I will be doing this. I'm about to switch from basically trying to build in private, like a lot of folks do, and that is to say, hey, we just build here, we just do the stuff and we don't show anybody what we're working on. ⁓ I've been talking with a few people. I've also been following a couple of other streams. And so I've been inspired by what, for example, Bridgemine has been doing. and he's building two tools for coders. Also, there's a few others that are building in public. There's actually three products that I wanna build in public on, and I'll be actually uploading those today. So by way of announcing, ⁓ I'm gonna post here. And each of these products ⁓ feature something that's useful. something that ⁓ I think a lot of people would be excited by as I am. The first is ⁓ open countries. And this is a system, and I've already been acquiring a lot of the data for this, ⁓ but this is a system designed to ⁓ help us understand a lot of the public data that's out there. You currently it's really hard to understand. You know, if you just go and download it and look at it, it's like, I don't even know what this is. And so until you can run it through some kind of a tool, ⁓ it just doesn't make human sense. Furthermore, a lot of that data needs to be mapped to something else. In other words, some of that data has to relate to somebody the other data before now you can make sense out of it. So, and what do I mean by all that? Well, Think about all the data that we have out there. School data, water quality data, ⁓ weather data. ⁓ Pick something. There's millions of data points ⁓ for each location throughout the U.S. So this open countries ⁓ will take any and all data that's publicly available from any location around the world. We'll start with the U.S. first and then... you know, help to visualize it. Now, something of this scale needs a lot of help. And I've got some talks with Snowflake underway. I just need to build some primitives before I can go and reach out to them and see as to how they can help, ⁓ you know, because they have actually some great tools for handling large data sets. But by way of this stream ⁓ and announcement, I'm inviting... anyone and everyone ⁓ to help us out on something like this. ⁓ Next, I have a little bit more of a fun one ⁓ inspired by, know, Pete Steinberger. And I've been talking with him up and on and I'll also try to jump in on his Discord and find out what their folks are doing ⁓ to try to manage this problem. But I've been working on it for about a week and a half and trying to fine tune the algorithms for it. But it's called ⁓ PR air traffic control ⁓ or pull requests, air traffic control. So the challenges in GitHub, even if you have a small project, you may end up with 50 different pull requests. And the question is, it becomes a of a combinatorics ⁓ and even exponential problem. because every piece of code that you add to your project, there's an inherent, there's a lot of inherent risk. ⁓ That code you add could break other things in your application and bad, you know, right now you're crashing. ⁓ Alternatively, that piece of code can add a security risk, ⁓ as has happened recently with Light LLM. ⁓ And even more than that, like a... Like I just said, there's 50 different pull requests. So if you stack them all together and try to just merge everything, you know, the whole thing becomes a bloody mess. So not good, right? So P-R-A-T-C is a way of managing that. And by managing, I mean, it will actually analyze each and every pull request. And I do mean all of them. And it would even do it in combination. It doesn't matter how many there are and in each case you might have 6,000 pull requests, you know, on a good day. So it will actually go through all of that. And there's also an Omni mode where it'll take the entire batch set and analyze all of it and recommend different pathways through all of that that would be beneficial to the project. It checks the security. It checks the crashiness as they call it. It checks, you know, all of it for you, automagically, as I like to say. I'm already in what I like to call 0.2. It's good enough for release. I'm going to probably hold back to security piece of it because, you you talk about crashiness. Yeah, it's crashing. So, you know, it kind of works. I mean, and then when you put it in Omni mode, it doesn't work. It's just like, well, I see why it's not working. So I want to go ahead and fix that before I, you know, add that to the project. But meanwhile, I can release everything else as it stands and y'all can take a look at that one. I have another one that's not quite Betty, ⁓ not even 0.1. So it's not that I'm holding back because I don't want to build in public. I do. It's just when I launch it, you know, I want y'all to see it ⁓ in the way that it's meant to be seen and right now they're just bad language here and there and everything else and it just wouldn't make sense. I just don't want to post that until it's right. So ⁓ those are the three that I'm working on that I can definitely talk about. I have two others ⁓ that I'm going to definitely demonstrate. ⁓ By the way these first, well two are ⁓ open source. One is cloud-based for now and only because I want to get it all working before it's fully released, but least build in public and invite commentary on it. ⁓ And then I have ⁓ a couple of closed-sourced projects that I will also be sharing ⁓ in the upcoming days. ⁓ And so, like I said, by way of this, I'm announcing my intent to build in public. I invite you all to go along with me for the journey. ⁓ You know, even when Bridge Mine started, he didn't have a single viewer. Just like, I don't know if even I even have one right now, but that's fine. You know, as we go, we will grow together. And I definitely want to help everyone here grow together with me. There's a lot of opportunity in all the projects. All the projects I just mentioned, there's a boatload of opportunities, they really are. And I invite you all along with me for the journey. Thank you.
درباره این اپیزود
Jefferson Nunn announces his new approach of building in public and shares details about his upcoming projects. He introduces 'Open Countries' as a tool to understand public data and 'PR Air Traffic Control' for managing pull requests. The focus is on open source projects and inviting collaboration from the community.
Takeaways
- Building in public
- Open source projects
Chapters
- 00:00 Announcing New Projects and Approach
انگلیسی
ایالات متحده آمریکا
رونوشت 🔗
جستجوی اپیزودهای گذشته
اپیزودهای قبلی Fast AI Development را جستجو کن.
سلب مسئولیت: پادکست و آثار هنری تعبیه شده در این صفحه متعلق به Jefferson Nunn است که متعلق به صاحب آن است و به Listen Notes، Inc وابسته یا تایید نشده است.
ویرایش
از کمک شما برای بروز نگهداشتن پایگاهدادههای پادکست سپاسگزاریم