Pitch an LLM Earnings-Call Analyst to an Equity Long-Short Team
Overview
What this challenge is about.
Pick 3 publicly available US tech earnings-call transcripts (from a free source like sec.gov filings or company investor-relations pages) and build a retrieval-augmented LLM workflow that produces a 1-page structured note per call: revenue surprises, forward guidance changes, sentiment shifts, and analyst Q-and-A red flags. Compare the LLM outputs to publicly available sell-side notes for the same calls. Define and report at least three honest accuracy metrics including a hallucination check. Wrap it in a 12-slide pitch covering scope, risks, governance, and a 90-day pilot plan.
The Brief
What you'll do, and what you'll demonstrate.
Show whether a retrieval-augmented LLM workflow can produce decision-grade earnings-call notes worth hiring an AI engineer to scale.
Earning criteria — what you'll demonstrate
- Design a retrieval-augmented generation (RAG) workflow for long financial documents
- Evaluate LLM outputs against domain ground truth with honest metrics
- Identify and mitigate hallucination risk in a high-stakes finance use case
- Pitch an AI product to a skeptical, results-oriented audience
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
AI Engineer
Standing up the retrieval pipeline plus the eval harness is the AI engineer's daily work in an applied LLM team.
This challenge sharpens
- retrieval-augmented-generation
- llm-evaluation
- prompt-engineering
AI Product Manager
Translating an LLM demo into a costed 90-day pilot plan for a portfolio manager is core AI PM craft.
This challenge sharpens
- financial-analysis
- stakeholder-communication
- llm-evaluation