Ian Robins<p>ML models don’t fail in the lab. They fail in production.</p><p>To build scalable, <a href="https://mastodonapp.uk/tags/observable" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>observable</span></a> ML systems, you need:<br>- MLflow for tracking<br>- Streamlit for monitoring<br>- Prometheus + Grafana for metrics<br>- SHAP for explainability</p><p>Learn how to make <a href="https://mastodonapp.uk/tags/ML" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>ML</span></a> work beyond the notebook: <a href="https://www.infoq.com/articles/building-observable-machine-learning-systems/" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">infoq.com/articles/building-ob</span><span class="invisible">servable-machine-learning-systems/</span></a></p>