Project #10 · High-Yield Production MVP
AI Interview Prep Coach (Mobile-First PWA)
Voice mock interviews for senior AI and full-stack roles. Speak your answer out loud, receive rubric-based scoring with quoted evidence, and refine STAR storytelling.
Ethical Guardrail: This tool evaluates interview communication structure and technical terminology. It does not predict hiring decisions or replace human evaluation.
Practice Terminal
Typed Fallback Ready2:00
Interview Question Prompt:
"How do you prevent hallucinations in high-stakes customer-facing LLM deployments?"
30 words
Rubric Evaluation
Composite Score:93/100
STAR Framework & Structure92/100
Evidence: "We implemented strict retrieval-augmented generation... and an offline golden benchmark..."
Clear Situation-Task-Action-Result flow with decisive implementation steps.
Technical Specificity & Depth96/100
Evidence: "below 0.65 similarity... verified document citations..."
Directly references threshold gating and retrieval verification.
Metric-Driven Evidence94/100
Evidence: "below 0.65 similarity... measuring groundedness and refusal accuracy continuously"
Cites concrete numerical guardrails rather than vague generalities.
Communication & Conciseness90/100
Evidence: Complete answer delivered in 34 focused words without filler.
Pacing is crisp, professional, and authoritative.