-
Our MVP should allow for data to be collected, sessions to be replayed, and experiments to be run.
[Open source (and self-hosted) alternatives to Hotjar & FullStory](https://posthog.com/blog/best-o…
-
As part of an experimentation me and @virgil are going to implement a clientside rewrite mode of tor2web.
This ticket deals with this experimentation that involve the following spec:
When accessing…
-
## Summary
Implement a Feature Flags management interface per service.
The motivation of this plugin is to either:
- Integrate with currently existing Feature flags management services such …
-
I've made a experimental EFI, more details @ [the-darkvoid issue](https://github.com/the-darkvoid/XPS9360-macOS/issues/174)
Only thing that doesn't work is WiFi, if we can get that to work we can …
-
Hello, I wanted to know if we can have a window to monitor the training process? Currently, I am saving videos during evaluations, but it slows down experimentation since I need to get the videos from…
-
By providing a hook that gets notified before and after the application starts, it's possible to support experimentation tracking.
Candidate backends:
- [mlflow](https://github.com/mlflow/mlflow)…
-
Je note ici les bonnes pratiques pour le benchmark à compléter.
- Utiliser le même itéré initial pour toutes les expérimentations que l'on compare entre-elles
- Passer les mêmes les options pour l…
-
As a SuperDuperDB user, I would like to connect to Apache Iceberg tables using Trino and/ or PySpark and apply AI models to these.
Since `superduperdb` supports `ibis` which supports both Trino and …
-
1. I want more freedom around choosing my main insight. Filters are missing. (Saved insights might be a way around this)
2. I want to rollout the entire experiment to a smaller % of selected people.
…
-
The goal is to understand how throughput is impacted when our system scales down & up
- This is to simulate node failure and recovery
One possible suggestion is to display a graph of throughput a…