A system to learn how to chat with dolphins, at https://www.dolphinwhispers.com and maybe become a contender for the 10 million $ Coller-Dolittle Prize https://coller-dolittle-24.sites.tau.ac.il/

I'm working on [Context Bonsai][1] - LLM harness tools that allow the LLM to prune messages out of the context, leaving behind a summary and keywords instead. In addition to a "prune" tool, there is a "retrieve" tool that allows it to recall the messages if needed.

In addition to these tools, I'm also building automation that will port the tools from the reference implementation (OpenCode) to other harnesses (Claude Code, Cline, Pi, Gemini, Kilo, Codex, others to come?). As well as automation that will either cherry-pick or re-implement commits onto the latest head from upstream.

[1]: https://github.com/Vibecodelicious/context-bonsai-agents#con...

[2]: https://blog.vibecodelicio.us/posts/how-i-fixed-context-wind...

A self hosted guestbook api that runs on my github pages site https://ianto-cannon.github.io/guestbook.html

A timelapse platform powered by community photos. The idea is to place a mount and QR code at fixed viewpoints around the neighbourhood. People scan, photograph the view, optionally add their name, and submit. Over time, the platform stitches those shots into a living record of how the place changes with seasons.

Just finished the software side using a boring technology and am about to order the materials for the first few locations. Curious to explore photo alignment once real submissions start coming in. Stitching all slightly different angled photos into a smooth animation seems interesting.

I have been quietly working for the last three years: a novel hierarchical and extensible modeling framework that can cleanly and efficiently embed any json-like object for any predictive modeling task with zero feature engineering.

json2vec enables users to, for example, build tabular / transactional foundation models like TabBERT / PRAGMA dynamically... by just declaring their data schema. This is a space in which Netflix, Stripe, Revolut, Capital One, Nubank, J.P. Morgan, NVIDIA, etc. have been developing for several years.

json2vec goes a step further from just tabular data or structured transactional data. It enables arbitrary structured "json-like" observations with hierarchical BERT-like transformer encoder blocks. Financial transactions, chess positions, flight itineraries, raw tabular data, rideshare activity, ecommerce, behavioral sequence models... Any raw data able to be represented in `json` can be encoded into a tree of embeddings, and used for downstream finetuning for supervised machine learning... No feature engineering required.

https://github.com/granthamtaylor/json2vec

json2vec supports extensible plugin support for new data types (numbers, categories, raw text, datetimes, hashable objects [think: IP addresses and phone numbers], and raw embeddings), all of which may be pretrained via MLM-like self-supervised learning. If your needs are not met with the built-in datatypes, the framework is extensible in that you may build your own custom datatypes (think: geographical coordinates). Built in decision heads for a subset of datatypes enable predictive modeling multi-task and multi-array outputs (predicting fraud at a per-transaction level, or a per-account level).

json2vec also enables built in data pipelines for 100b+ training observations streaming from cloud storage. These pipelines integrate with layer of programmatic data querying and UDFs can consume the vast majority of upstream data processing so that developers don't waste time on massive batch data preprocessing jobs before model training.

Oh, and the best part: the model architectures instantiated by json2vec are mutable. Model developers can add and remove features and targets at their whim - allowing for truly reusable foundation models that can adapt for each individual use case.

My hope is that with a standardized hierarchical modeling framework, interested organizations can better collaborate with one by sharing reusable logic with one another instead of hardcoding use-case-specific architecture.

I am finally getting close to my vision of `write once run everywhere with SwiftUI`. The idea is to create a drop-in replacement called OmniUI which will have different renderer backends (I currently have TUI w/ notcurses and Adwaita/GTK working)

s/import SwiftUI/import OmniUI/

As long as you aren't using Apple platform specific libraries like Vision, you should be good for the most part. I am going to make my Gopher browser (https://web.navan.dev/iGopherBrowser/) the first target. I have done some extra stuff like reimplementing CoreData/SwiftData to make it work on Linux.

I am going with Adwaita instead of pure GTK because I like the opinionated approach they have with their design language. I think the reason SwiftUI works is because you can get pretty looking apps without thinking too much.

Projects like adwaita-swift, and swift-cross-ui do exist, but I want my library to be a drop-in replacement. I don't want to be inspired by SwiftUI, I want to use SwiftUI everywhere!

Social network for gamers (iOS/Android, soon web).

Tinder meets Discord and, somehow, they have their way with Uber/Calendly.

It's live if you want to test it: https://jynx.app/

Let me know what you think of it. The main goals I want to achieve are: 1. help with social isolation 2. help e-sport team with sourcing and organizing

Working on a way for us to communicate software changes properly with people, systems and AI.

I'm frustrated at how we just do library updates and get whatever is next. Things break. Finding a changelog sucks. Vendor updates maybe put something in X.

We can do better so I'm starting an open initiative to tackle that.

I've made https://rankr.click

It's a little web application that allows for the ranking of all kinds of abstract entities. Think of the merging of Goodreads for books, Vivino for wine, Letterboxd for film, etc. This will allow you to instead rank whatever you want across a variety of different categories in a single place.

Using your rankings across all these different fields, you can draw analysis of what you like, and in future I'd like to add a little personal (not an ad) recommendation engine to help you find new stuff based on your actual interests across loads of different categories.

From a technical point of view, its been a great learning opportunity on how to fully host a complete stack using an opiniated, but cross-platform orchestrator, allowing me to host this wherever (bare metal VPS, homebrew system, cloud provider) in a flash.