Cookies

With your permission, we use Google Analytics cookies to measure visits and improve the site. Necessary cookies are always on. You can change your choice or withdraw consent anytime in the footer, under Cookie settings. Cookie Policy

Skip to content

Advanced Agents ·Intermediate

Earnings Call Analyst Agent

About this app

An investor-grade earnings call companion that turns any YouTube earnings call into a playback-synced analyst workspace. Paste a call URL, watch the video, and let ADK agents surface the numbers, tone shifts, filing context, and market-moving surprises that are easy to miss in a live call.

  • Complete open-source code
  • Step-by-step tutorial

Technical sheet

Category
Advanced Agents
Difficulty
Intermediate
Models
Gemini
Frameworks
Google ADK
Runs locally?
Needs cloud APIs
Required keys
GOOGLE_API_KEY

License Apache-2.0 · content curated and maintained by Doutor AI.

Full tutorial

An investor-grade earnings call companion that turns any YouTube earnings call into a playback-synced analyst workspace. Paste a call URL, watch the video, and let ADK agents surface the numbers, tone shifts, filing context, and market-moving surprises that are easy to miss in a live call.

This is built for the real earnings workflow: instead of reading a transcript after the fact, you can follow management commentary with an agentic research layer that keeps every insight tied to the quote that triggered it.

📡 Earnings Call Analyst Agent architecture

Features

Agentic Call Research

  • Identifies the company, ticker, fiscal period, and peer set from the YouTube metadata and transcript opening
  • Builds a research pack with SEC filings and current market context
  • Uses an ADK news agent with Google Search grounding for current market context
  • Hides unresolved context instead of showing empty research panels

Quote-Anchored Signal Detection

  • Creates analyst cards only when the transcript contains a real investor signal
  • Anchors every card to the exact quote and timestamp that triggered it
  • Filters out greetings, safe-harbor boilerplate, and generic upbeat commentary
  • Reveals cards as playback reaches the relevant moment in the call

Earnings Intelligence Cards

  • Flags financial metrics, margin pressure, guidance language, demand commentary, pricing, cash flow, and capex signals
  • Separates company-specific statements from peer or sector context when evidence is available
  • Calls out CFO hedging, confidence shifts, defensiveness, and unusually specific language
  • Adds compact tables or chart summaries only when they clarify the finding

Caption + Audio Resilience

  • Uses YouTube captions when available for precise timestamps
  • Falls back to ADK-powered audio transcription for captionless videos
  • Realigns generated cards to the closest caption segment so the video and quote stay in sync
  • Keeps the transcript, research pack, and analyst cards tied to the same source timeline

How to get Started?

This agent lives in advanced_ai_agents/single_agent_apps/earnings_call_analyst_agent.

  1. Clone the GitHub repository
git clone https://github.com/Shubhamsaboo/awesome-llm-apps.git
cd advanced_ai_agents/single_agent_apps/earnings_call_analyst_agent
  1. Install the required dependencies:
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
  1. Configure Vertex AI or Gemini API key:
cp .env.example .env

For Vertex AI / Google Cloud auth:

GOOGLE_GENAI_USE_VERTEXAI=True
GOOGLE_CLOUD_PROJECT=your-google-cloud-project-id
GOOGLE_CLOUD_LOCATION=global

For Gemini API key auth:

GOOGLE_GENAI_USE_VERTEXAI=False
GOOGLE_API_KEY=your-google-api-key
  1. Run the FastAPI app:
PYTHONPATH=.. python -m uvicorn earnings_call_analyst_agent.live_demo.server:app --host 127.0.0.1 --port 4188
  1. Open the app:
http://127.0.0.1:4188

Paste a YouTube earnings call URL. The app builds the research pack first, then reveals analyst cards as the video reaches each quote.

From the same category

Advanced AgentsIntermediate

AI Agent Governance - Policy-Based Sandboxing

Policy-Based Sandboxing: Define what your AI agent can and cannot do using declarative policies - Action Interception: Catch and validate agent action…

  • OpenAI
Advanced AgentsIntermediate

AI Consultant Agent with Google ADK

A powerful business consultant powered by Google's Agent Development Kit that provides comprehensive market analysis, strategic planning, and actionab…

  • Gemini
Advanced AgentsIntermediate

AI Customer Support Agent

This Streamlit app implements an AI-powered customer support agent for synthetic data generated using GPT-4o. The agent uses OpenAI's GPT-4o model and…

  • OpenAI

Need a tailor-made solution?

Not sure where to start with AI in your business? Get a free assessment and a tailored plan - no commitment.

Save hours every week with AI, no hassle

Doutor AI gives you 1,859 ready-to-copy AI prompts, by industry and in plain English - plus a complete technical library for those who want to go further.

See plans