Building Your First AI Agent
Practical Framework, Tools & Tips
Press space or right arrow to begin
n8n visual workflow + AI = Beautiful results
Drag & drop nodes
Complete workflow
Smart planning
Professional HTML
Streamlit + OpenAI Agents SDK
Made in the AI Agent Bootcamp
Intelligently routes user queries to specialized agents, combines results, and maintains conversation context
Searches internal knowledge base using vector similarity, retrieves company documentation and policies
Performs real-time internet research for current events, news, and external data using SerpAPI
Creates and formats professional reports in Google Docs, handles document generation
Hierarchical Architecture
PGVector RAG
Live Web Search
Google Docs API
Built in the AI Agent Bootcamp
investment-agents/ ├── research_agents/ │ ├── portfolio_manager.py │ ├── fundamental_analyst.py │ ├── macro_analyst.py │ └── quantitative_analyst.py ├── tools/ │ ├── fred_api.py │ └── memo_editor.py ├── prompts/ │ └── *.md (agent instructions) ├── database/ │ ├── models.py │ └── memory.py ├── streamlit_app/ │ └── app.py └── main.py
@function_tool
async def run_all_specialists_parallel(
query: str
) -> str:
"""Run 3 agents simultaneously"""
tasks = [
Runner.run(fundamental_agent, query),
Runner.run(macro_agent, query),
Runner.run(quant_agent, query),
]
# Execute in parallel - 3x faster!
results = await asyncio.gather(*tasks)
return combine_analyses(results)4 Specialist Agents
Parallel Processing
PostgreSQL + Vector
Heroku Deployed