Creating a Researcher in Python with OpenAI

Overview

Learn to build the control logic for a multi-step AI research assistant. Structure the main python script to manage input, control iteration, and orchestrate modules for query generation, relevance checks, context extraction, and report creation.

Implement an LLM-driven workflow for iterative search and comprehensive report generation.

Syllabus

• Unit 1: Setting Up the Basic Structure for DeepResearcher
Building the DeepResearcher Framework
Implementing the Main Research Workflow
Adding Error Handling to DeepResearcher
• Unit 2: Generating Search Queries with OpenAI
Designing Prompts for Search Generation
Generating and Parsing Search Queries
Validating AI Responses for Robust Research
Testing Your Search Query Generator
• Unit 3: Parsing and Selecting Useful Information
Collecting Web Content for Research
Creating Prompts for Content Relevance Evaluation
Evaluating Content Relevance with AI
Designing Prompts for Information Extraction
Extracting Gold from Web Content
• Unit 4: Iterative Search: Refining Research Through Multiple Rounds
Building the Iterative Research Loop
Accumulating Knowledge Across Research Rounds
Creating the Research Planner Brain
Safely Parsing LLM Query Responses
Testing Your Iterative Research System
• Unit 5: Creating the Final Research Report
Crafting Prompts for Research Reports
Implementing the Final Report Generator
Print the Final Report
Testing Your AI Research Assistant

---

Subjects

Computer Science