Hacker News
- JSON is all you need: Easily monitor LLM apps with structlog https://ploomber.io/blog/json-monitor-llm/ 0 comments
- How good is GPT-4o at generating Flask apps? Surprisingly promising https://ploomber.io/blog/gpt-4o-flask/ 0 comments
- Who needs MLflow when you have SQLite? https://ploomber.io/blog/experiment-tracking/ 110 comments
- How good is GPT-4o at generating Flask apps? Surprisingly promising https://ploomber.io/blog/gpt-4o-flask/ 7 comments flask
- Deploying vLLM: a Step-by-Step Guide (to host your own ChatGPT) https://ploomber.io/blog/vllm-deploy/ 9 comments selfhosted
- Continuous Integration for Data Science https://ploomber.io/blog/ci-for-ds/ 7 comments datascience
- A systematic overview of prompt engineering https://ploomber.io/blog/prompt-engineering-techniques/ 9 comments datascience
- JupySQL: Better SQL in Jupyter https://ploomber.io/blog/jupysql/ 3 comments python
- Scaling notebooks in the cloud with zero infrastructure configuration https://ploomber.io/blog/cloud-notebooks-gui/ 12 comments python
- Which Pandas version did I use? Recover dependency versions from legacy notebooks https://ploomber.io/blog/reproducible-environments/ 6 comments python
- [E] Selecting the optimal threshold for a binary classifier https://ploomber.io/blog/threshold/ 9 comments statistics
- [D] Can I trust my model's probabilities? A deep dive into probability calibration https://ploomber.io/blog/calibration-curve/ 17 comments statistics
- How do you deal with Jupyter notebook debt? https://ploomber.io/blog/nbs-production/ 61 comments datascience
- Why (and how) to put notebooks in production https://ploomber.io/posts/nbs-production/ 4 comments datascience