It's more of a fun project right now, but hoping to unearth some marketable use cases and ideas as I go along. I'll do a proper launch in a month or two.
Fun to build, mostly just thinking through the UI, catching up with all the new AI models coming out these days. Would be a ton of work to really polish and monetize an iOS / Android app like this, not interested in taking it much further.
• i rewrote a giant piece of software from scratch and will demo it for the owners of the product tomorrow (wish me luck)
• creating an mvp for my new startup that will tackle the AI alignment problem (looking for pre-seed investment!)
• polishing a compiler for a programming language i made
• writing essays on productized services, selfish recycling, elm, and wolfram alpha
• searching for a healthcare plan
• training for the aids/lifecycle (7-day bike ride from SF to LA)
[3] hello@taylor.town
At the moment working on functionality to easily ingest email communications with recruiters/companies so that it can truly be a system of record (without needing to give access to your inbox or any of that).
Firestarter https://github.com/mcapodici/firestarter, is perhaps a more humble project than most here, but happy if anyone is interested to follow it.
It is an attempt to build what is (in my opinion!) a great stack for quickly iterating on a web app idea. Probably about a month off from first release, which will be a simple TODO app you can sign up for, with full instructions on getting it set up and deployed that would be about an hour's work. From that point you have a semi-professional setup with devops and CI taken care of so you can move faster on your idea or startup. No need to touch a server (there isn't one to touch!)
Uses Firebase/NextJs/Vercel/React/Tailwind/Github Actions/Jest/Testing Library
I chose firebase because it is a very good choice for getting up to speed fast. The downside it is a lock-in, and not open source. However the code is written in such a way that swapping out to Parse, for example would be fairly easy to do.
It would be great to not start from scratch everytime and still have base ui that could be customizable.
Codepen partially solve this, but searching them takes time and often they are not customizable.
There are multiple sites but they just redirect to selected codepen.
Don't know if it's viable. I am banking on the rise of indie devs building products without designers.
Currently looking for a cofounder so feel free to reach out :) https://ellispinsky.com/
This will help creators build with AI using a cool UI with tons of features. Initially for images, then I'll expand to documents and code. I'm aiming to launch in the next few weeks.
https://github.com/onlineornot/onlineornot
I'm finding building a CLI fun because it's still a "frontend", but you have to think harder about how the UX should work - since you can't just make it pretty.
(it's also a cheap way to grow the roadmap for my public API since I need to build all these endpoints to make what I want to do possible)
It's a fun side project but something I think needs to exist. I found it difficult to see how places were changing over time (ie are schools getting better or worse, is income increasing or decreasing, etc..) Would love to hear your feedback or meet individuals interested in working together. hello@searchaway.co
think:
- fixes around the house / todos
- image as a reminder
- wishlist (some photo lists can be public and shareable)
- recommendations (when someone is showing you a book or sth, just capture and save it)
I'm still at an early experimentation stage, but have I'm documenting progress at https://twitter.com/visuelapp
Its a simple tool that plots how data changes in a google sheet over time. So the categories of value per row, number of rows added & removed and some more complex things like averages.
I use it personally to monitor user and finance related sheets that others are editing so I can see how things are changing over time.
In February, I was working on two projects and just released them both recently:
https://trackmybot.com - public ChatGPT responses with public archive.
https://artsy.sh - dynamic drawing tool featuring DALLE-2 and Stable Diffusion 2.1 engine models.
Onboarding and messaging tool for small SaaS businesses. Should be ready in a week.
This will slowly evolve into a time-to-value helping tool, which means clients will be able to help their users minimise time-to-value of their product, thus reducing churn and improving free to paying conversions.
First is Concise Encoding, a secure, binary and text ad-hoc data format with full type support: https://concise-encoding.org/
Example of the text format:
c1
{
"string" = "Strings support escape sequences: \n \t \[1f415]"
"boolean" = true
"binary int" = -0b10001011
"octal int" = 0o644
"decimal int" = -10000000
"hex int" = 0xfffe0001
"very long int" = 100000000000000000000000000000000000009
"decimal float" = -14.125
"hex float" = 0x5.1ec4p+20
"very long flt" = 4.957234990634579394723460546348e+100000
"not-a-number" = nan
"infinity" = inf
"list" = [1 2.5 "a string"]
"map" = {"one"=1 2="two" "today"=2020-09-10}
"bytes" = |u8x 01 ff de ad be ef|
"int16 array" = |i16 7374 17466 -9957|
"uuid" = f1ce4567-e89b-12d3-a456-426655440000
"date" = 2019-07-01
"time" = 18:04:00.948/Europe/Prague
"timestamp" = 2010-07-15/13:28:15.415942344
"null" = null
"media" = |application/x-sh 23 21 2f 62 69 6e 2f 73 68 0a 0a
65 63 68 6f 20 68 65 6c 6c 6f 20 77 6f 72 6c 64 0a|
}
Almost all existing data formats are horribly insecure and ripe for exploitation (many of them are already). And they can't be fixed because the formats aren't versioned.-----------------
But describing the data format was painful because most existing metalanguages are only for describing textual grammars, and the few that support binary aren't really suited for documenting for human consumption.
So I'm putting the finishing touches on Dogma ( https://github.com/kstenerud/dogma/blob/master/dogma_v1.md ), a modernized BNF-style metalanguage with better expressiveness and binary grammar support.
For example:
A UDP packet consists of a source port, a destination port, a length, a checksum, and a body. The length field refers to the size of the entire packet, not just the body.
udp_packet = src_port
& dst_port
& uint(16, var(length,8~) )
& checksum
& body(length - 8) # subtract header from length to get body length
;
src_port = uint(16,~);
dst_port = uint(16,~);
checksum = uint(16,~);
body(length) = uint(8,~){length};