- a distributed simplex downlink protocol for low power satellites. Think torrenting for LEO
- my startup, which is building a closed loop artificial pancreas for hospitals. First human trials are coming up early this year!
101–110 of 1001 posts
- a distributed simplex downlink protocol for low power satellites. Think torrenting for LEO
- my startup, which is building a closed loop artificial pancreas for hospitals. First human trials are coming up early this year!
The autolayout algorithm will be used for generating pretty software architecture diagrams from text that get you 90% of the way there, and then you can tweek it to perfection via a UI.
I have an alpha out of the algorithm on https://terrastruct.com and it's by far the hardest thing I've worked on, and it mostly works, though I'm constantly finding ways to improve it.
[0] It needs to handle containers and clusters, its connections should be mostly orthogonally routed (the tree structure with curved routing in default graphviz is mostly unsuitable), prefer symmetrical structures while reducing total edge distance, etc.
A web application that teaches Software Engineering Management theory
Me and a friend have been working on a game that I describe as "Factorio but for programming". Essentially you're an AI who's mission is to mine every resource from a planet. To keep things efficient you start out small and have limited processing power. You're only able to run an assembly like language that we've developed. As you get more resources you can research new things like functions, variables, type argumen…
(Yes, FTUE needs a lot of work, working on that now!)
HyperTag helps humans intuitively express how they think about their files using tags and machine learning. Represent how you think using tags. Find what you look for using semantic search for your text documents (yes, even PDF's) and images. Instead of introducing proprietary file formats like other existing file organization tools, HyperTag just smoothly layers on top of your existing files without any fuss.
The general-purpose search engine (which is built on top of Bing) is however not ready for daily use, because I haven’t figured out how to cover the costs of Bing’s API.. so after a certain amount of monthly searches then it will redirect all queries to Google. DDG and Ecosia appear to be relying on Microsoft Advertising, but not sure how to join that, especially when the website has a very limited user-base. Perhaps requiring a paid subscription for the general-purpose search is the best option for now.
The initial plan is to launch different categories, such as Podcast, Food, News, Job, Shop, Flight, Code, Sport, etc. since this will allow us to provide superior search results compared to any general-purpose search engine.
So far we have implemented support for the categories Search, Math, Currency, as well as the recently launched Domain name finder (https://ask.moe/domain).
If anyone is interested in working together to obtain and maintain high quality data for these categories then feel free to reach out (but I’m quite busy these days so expect a slow reply) :)
I’m also working on a gaming/esports website (but not ready for launch), as well as the occasional weekend projects.
It runs a machine learning model in your browser to convert the text into points in a high dimensional space, and then it projects those points down to 2/3D.
Right now you can tell it to visualize post titles or comments from any subreddit or load an hourly updating snapshot of Twitter.
You can also view your own data in it by selecting the New Nebula option. The data never leaves the browser, which also means the ML models are run in-browser (via tensorflow.js). This part might be slow and only works in Chrome unfortunately.
If you're interested in this kind of thing, I'd love to hear from you! Here or by email (grady.hsimon at gmail)