Generative AI + Python: Unlocking Efficiency, Personalization, and Insight
summary
Three Streamlit apps demonstrate concrete results: a data‑analysis assistant that writes pandas and Plotly code, trains logistic/regression random forest and SVC models, and surfaces a 26% churn overall and 54% churn for month-to-month electronic check customers; a brief generator that turns structured customer rows into tailored emails and product recommendations; and a collaborating‑agents system (GPT‑4o, Claude 3.5 Sonnet, Llama 3 170b via Grok) that runs multi‑round dialogues, assigns tasks, and emits a strategy plan. Code execution retries up to five times and OpenAI API integration are shown working end‑to‑end.
AI-written summary · about
chapters (14)
- 0:00Welcome
- 1:03Speaker background
- 2:04What is generative AI
- 3:07Democratizing data analysis
- 4:56Demo: Data analysis app
- 12:34Behind the scenes of tool
- 15:46Personalizing with AI
- 17:09Demo: Brief generator
- 19:42Designing brief generator
- 21:09Improving collaboration
- 22:36Demo: Agents app
- 28:42Designing multi-agent system
- 29:44Key insights and future
- 30:49Q&A
related talks
40mCreating Interactive GenAI applications powered by Streamlit, open source LLMs
58mPractical Tips for Using Generative AI in Data Science Workflows
45mBringing the Power of AI to Your Application
40mDavid Fogl: Leveraging Generative AI for Enhanced Software Testing
21mRevolutionizing software creation: Harnessing the power of GenAI
34mBuilding Multi-Agent Generative-AI Applications with AutoGen
36mUnleashing Generative AI: Improving Developer Productivity in SRE
45mBuild Agentic AI Apps with the Autogen framework