🤝 Open to Collab

Andrei Ovcharenko

Kreout

AI & ML interests

Java backend development, Spring Boot, Kafka, infrastructure and automation. Interested in local AI tools and practical developer workflows. Open-source developer tools: https://github.com/krotname.

Recent Activity

upvoted an article about 9 hours ago

We Fine-Tuned a Memory Classifier for Our Agent. Here's Why We Didn't Ship It

reacted to Baekpica's post with 👍 about 9 hours ago

Does anyone have this problem? I'm seeing a sudden, highly unusual spike in download statistics for one of my quantization releases, Baekpica/Solar-Open2-250B-Mixed-Quant-GGUF. https://huggingface.co/Baekpica/Solar-Open2-250B-Mixed-Quant-GGUF It had been increasing by roughly 2K reported downloads per day until I enabled the gated access user settings. It was in a fairly linear pattern, despite there being no corresponding increase in likes, community activity, external exposure, or apparent popularity or hypes of the upstream base model. It's very weird… Please tell this to HF maintainers..! I cannot reach them even with my email report.🥲

reacted to uuhugg's post with 👍 about 9 hours ago

A retry can succeed and still be unjustified. RetryBench separates whether an agent’s retry is supported by the visible contract from what happens afterward. After a timeout, success alone does not tell us whether retrying was justified. Try the interactive demo (English / 中文): https://huggingface.co/spaces/uuhugg/retrybench-demo Read the findings and reproduction guide: https://huggingface.co/blog/uuhugg/retrybench-retry-permission How do you evaluate retries in your agents?

View all activity

Organizations

upvoted an article about 9 hours ago

Article

We Fine-Tuned a Memory Classifier for Our Agent. Here's Why We Didn't Ship It

javadtaghia

•

4 days ago

• 1

Andrei Ovcharenko

AI & ML interests

Recent Activity

Organizations

Kreout's activity

We Fine-Tuned a Memory Classifier for Our Agent. Here's Why We Didn't Ship It