Multi-language Data Wrangling and Acquisition Conversational Agents
summary
Compares concrete data-wrangling examples across Julia (DataFrames), Python (pandas), R (base and tidyverse) and Wolfram Language, and describes designing conversational agents that map natural-language commands into code for Julia, Python, R, SQL and WL (using WL’s external evaluator). Then presents a CA design for data acquisition workflows based on a finite-state machine, covering metadata taxonomies, dataset recommender/search ideas, ingredient-variable and introspection queries, random-data and obfuscation specs, and notes extensions for ML model acquisition, with implementations for R and WL.
AI-written summary · about
related talks
31mGenerative Conversational AI Interoperability
59mMaking the most of artificial and human intelligence for data science
34mProgramming Languages for AI Programing Agents
30mRyoma: Building Open Source AI Agents for Enterprise Data Analysis
47mBuilding Python based AI Systems with LLMs
20mIris: A Conversational Agent for Complex Tasks
48mDyalog and AI
26mWelcome to the AI Jungle! Now What?