Just my two cents. I'm on the Plus plan, I ask gpt-5.6 sol / high to analyze a vibe-coded codebase (~50k LoC) and write a plan to make it production ready. It wasn't a great prompt, I just wanted to test it quickly. It ran for ~15min and consumed 95% of my 5h quota (I thought it was gonna crash). The output is excellent but just a heads up that it consumes a lot of quota!
Ok after re reading I think you have issues with long running spans, I think you should break down your spans in smaller chunks. But a trace can take many hours or days, and be analysed even when it’s not finished
I’ve implemented OTEL for background jobs, so async jobs that get picked up from the DB where I store the trace context in the DB and pass it along to multiple async jobs. For some jobs that fail and retry with a backoff strategy, they can take many hours and we can see the traces fine in grafana. Each job create its own span but they are all within the same trace.
Works well for us, I’m not sure I understand the issue you’re facing?
Did you consider also caching the coordinates returned by moondream? I understand that it is cheap, but it could be useful to detect if an element has changed position as it may be a regression
Do you know when we can expect an update on the realtime API? It’s still in beta and there are many issues (e.g voice randomly cutting off, VAD issues, especially with mulaw etc…) which makes it impossible to use in production, but there’s not much communication from OpenAI. It’s difficult to know what to bet on. Pushing for stt->llm->tts makes you wonder if we should carry on building with the realtime API.
Call it “stalled” if you like, it’s stable and it’s pure joy. I can just get stuff done with Clojure. And things that may seem inactive, like that lib that you need that hasn’t had a commit in 8 years, turns out that it just works and doesn’t need to change. This is commonplace in Clojure.
Spec is still alpha and I’m not sure it will evolve more or if it’ll be something completely different. At least they’re not pushing you down the wrong path. Use/look at Malli instead of spec.
Could you also pick up the audio announcements? Could be weird for the app to request microphone permission though, but I think lots of lines have them and should be doable to transcribe them locally on the phone. I think your approach is better and simpler, but as I was reading the edge cases (like someone changing train and going back in the other direction), the audio could maybe help getting from 90% accuracy to 98%. Sounds a bit more involved to implement though.
Honest question, what’s wrong with docker desktop? Looking at all the alternatives suggested it’s not clear to me why any other tools are better? I’m not using k8s locally, just docker compose. To connect to our remote k8s cluster, I use IntelliJ k8s extension (I just need to do some basic dev tasks, I’m not administrating the cluster)
t dot h at sent dot com