AI/ML Proof of Concept Project Manager Simulation — Google

The scenario

Google's Talent Acquisition team processes over 3 million applications per year across 170+ offices worldwide. Recruiters spend an estimated 40% of their time on initial resume screening — reading, sorting, and shortlisting candidates before any human conversation happens. Despite Google's reputation for rigorous hiring, the screening process is inconsistent: two recruiters reviewing the same resume pool will produce different shortlists 35% of the time. Time-to-fill for engineering roles averages 62 days, and hiring manager satisfaction with candidate quality sits at 71% — both metrics leadership wants to improve. The AI-Enhanced Talent Acquisition Initiative was proposed by the VP of Talent Acquisition to explore whether machine learning can augment (not replace) recruiter decision-making. The concept: an AI system that scores and ranks candidates based on job requirements, historical hiring data, and structured competency signals — giving recruiters a prioritized shortlist instead of an unranked pile. The system must be explainable (recruiters need to understand why a candidate was ranked) and bias-tested (Google's Responsible AI principles and EEOC guidelines are non-negotiable). Before committing to a full platform build, leadership approved a $380K proof of concept. Your job is to evaluate whether the technology works in practice — not in a lab. Two pilot teams (Engineering hiring in Mountain View and Sales hiring in New York) will use the system for 4 weeks of live recruiting. The PoC must answer three questions: Does AI screening improve recruiter efficiency? Does it improve candidate quality as measured by hiring manager satisfaction? And can it do both without introducing demographic bias? The answer to these questions will determine whether Google invests $5M+ in a global rollout.

What you'll do as the project manager

• →Evaluate whether AI-powered candidate screening reduces recruiter screening time by at least 30% across both pilot teams
• →Measure whether AI-ranked shortlists improve hiring manager satisfaction scores from the current 71% baseline to 80%+
• →Complete bias testing across 6 demographic dimensions (gender, race, age, disability, veteran status, national origin) with zero statistically significant adverse impact
• →Deliver a PoC recommendation report with a clear invest / don't invest / invest-with-conditions recommendation by Day 25
• →Achieve 85%+ recruiter adoption rate among the 12 pilot recruiters during the 4-week evaluation period

Project management skills you'll build

The challenges you'll navigate

• •Bias in AI model — if the model learns from historically biased hiring data, it may replicate or amplify demographic disparities
• •Recruiter adoption — pilot recruiters may resist the AI system if they feel it threatens their judgment or autonomy
• •Scope creep — sponsor enthusiasm may push the PoC beyond its evaluation mandate into premature production features
• •Vendor dependency — Eightfold AI's integration timeline and model performance are partially outside Google's control
• •Evaluation credibility — if the PoC isn't rigorous enough, skeptics will dismiss the results regardless of outcome

Technology & stakeholders

You'll manage 6 stakeholders, including Anika Desai (VP, Talent Acquisition — Global), Nina Patel (Senior Program Manager, People Ops PMO), Marcus Chen (Staff Engineer, Talent Acquisition Platforms), and more.

What you'll walk away with

A verified, shareable record of a completed enterprise project — plus the PMO deliverables you produced along the way (charter, project plan, SteerCo deck, closure document). It's real, demonstrable project management experience you can put on your resume and speak to in interviews.

Frequently asked questions

Related simulations