Django for AI: Deploying Machine Learning Models with Django with Will Vincent
summary
Shows a concrete pipeline: Iris dataset, SVM, scikit-learn, joblib (97% accuracy) trained in ~20 lines in a notebook, saved and loaded by Django views, forms and the ORM to record predictions and retrain, deployed on a PaaS. Contrasts that with LLMs which need GPUs, huge tokenized corpora and heavy inference, and demonstrates a local setup: Olama running Gemma 34B streaming tokens into Django via StreamingHttpResponse and HTMX. Practical demos and repo links are provided.
AI-written summary · about
chapters (7)
- 0:00AI and Django intro
- 4:16FastAPI vs Django trend
- 8:50Classic ML with Django
- 19:22How large language models differ from ML
- 26:18LLM training and inference
- 34:24Streaming AI chatbots
- 37:12Django's AI future
related talks
28mHere Come The Robots - Django and Machine Learning
37mSoftware Development for Machine Learning in Python
41musing-django-docker-and-scikit-learn-to-bootstrap-your-machine-learning-project/
29mBuilding Machine Learning demos with Python
33mYour best Bet: Effortless MLOps with Python Models in
26mBuilding and Scaling an AI Startup with Async Django
27mDeveloping and Deploying Models at Scale — without Spark!
33mDeploying and Scaling Spark ML and Tensorflow AI Models