Description on the tin.

I found it hard to get fares and timetables for the Elizabeth Line in London between X and Y quickly hence I built this as an experiment using AI tooling.

Clusters ready to deploy you applications on. It is meant for companies who want their own private infra, but do not want to maintain it. It is multi cloud, with a focus on europe.

Our secret weapon is “Menno”: an assistant which helps you setting up the cluster and deploying applications. We use the assistant for handson demo’s and workshops and got some really good feedback on it.

The web spec is vast and endless, if you try to build something actually works for modern sites, it will be huge. If you want to keep it lightweight, it won't work for most sites.

Just use chrome-headless-shell or webkit. Don't reinvent the wheels.

And also Kronikle https://www.kronikle.eu/en/ an app for local libraries. It creates public display interfaces that (I think) are good for showcasing their events.

- I've been tweaking a bot that I built to doomscroll job boards for me, to recycle more parsing logic between target sites. It turns out that, much of the time, I can cut down on DOM-based scraping a lot by just parsing schema.org-based JSON-LD. Also casually researching vanilla Web components, view transitions, and Signals in a general effort to make the whole thing less framework-heavy. (Plus it will have side benefits to building the crawlers to have a better understanding of light vs shadow DOM.) Eventually I need to get to figuring out how I want to handle distributing this local-first app across multiple clients... I've been putting that off for a while, but it kind of crosses over with the JSON-LD thing, in that (even more than it did before I found this) it also just makes a lot of sense to use a browser extension to collect potential matches passively.

- 3D printing experiments. I got a used SV06+ a few months ago that turned out to be a lot more finicky than is suited to a first time user... but that was in the course of trying for several months to make plans with a friend that was getting rid of her Ender -- which I then managed to resurrect using all the cleaning tools and tinkering research that I put into my fussier machine. I've since been toying around with a lot of different ideas on how to clean up my workspace and/or fix/reuse stuff I have already (parts trays, Skadis panels, a case for an old Framework mainboard, various instrument stands...) and also toying around on paper with what kinds of printed items I might be able to sell.

- Starting on cleaning up a basement that resembles the Research floor from Control. Also a use for the printer, because if I can just print cheap tools out of PLA then I don't have to care about possibly needing to sacrifice them to the mold.

- Various Coursera stuff. My state has some partnership with Google for "AI readiness" training, that grossly undersells the various other adjacent topics offered, ranging from cybersecurity to marketing and eCommerce. Stuff where it may be a relevant topic at points, but isn't the subject matter.

I wonder if anyone reads these? If you do, if you want to complain to me about observability, please shoot me an email at hello@telemetrymachine.com!

We have worked in the observability industry for a combined 15 years and we are trying to right the wrongs we have seen in the industry. It seems like the existing companies are more interested in extracting more and more money from giant customers than actually helping people run their software well.

I won't bury the lede: what we've built is OTel native, wide events native, and works well with agents. We are also, shamelessly, cheaper than everyone else. One of the reasons observability is struggling with adoption is because it's too damn expensive.

So that's what we're building at telemetrymachine.com: an observability product focused on helping you actually track your software (starting with OTel traces and logs). We are really inspired by Observability 2.0/wide events, which really just is an invitation to stuff as many attributes, with as much cardinality as you want, into your spans/logs.

One thing we are also doing differently is trying to be as lean as efficient as possible. We are just two people now, and we are going to be very careful about hiring anyone. Most importantly, what we need to do is avoid a giant salesforce as much as possible. I wrote about this in a blog post[1] but the reason observability is so expensive is actually the giant sales force these companies tend to have.

We also put a lot of work into actual infra efficiency – ie we run on bare metal, and wrote a state of the art[2] storage engine. All of this is to help get prices down while still building a sustainable business.

We are very keen to engage with observability users, so if you do, or do NOT, use observability in your services, please reach out and we'd love to hear your problems and thoughts!

[1] https://telemetrymachine.com/blog/observability-doesnt-have-... [2] https://telemetrymachine.com/blog/the-new-state-of-the-art-i...

Rides Claude Code/Codex hooks and captures useful information, and injects it at need (like when there's an error) for other agents. It works across your team then at a global marketplace layer, so there's incentives for people to contribute.

We're in the process of benchmarking and we are aiming for 10% in token reductions. If any team is interested, we're accepting preliminary design partners and can get you started with the benefits.

It's based on the Server-Driven UI philosophy where the backend streams both code and UI into a frontend that just knows how to render what it's told. This is pretty cool because it makes it makes over-the-air updates possible.

It also follows the batteries-included philosophy of Rails. Has an ORM, background and cron jobs (all based on Postgres), Django-style migrations, JSX Emails and pretty great FFI that's built around a simple `"use native"` directive, in the React Server Component fashion).

https://sunred.app/