💾 Auto-Save: Your work is automatically saved as you type. You can close the browser and come back later.
Week 3 · Student Packet
Pitch Coaching Guide & Peer Feedback Sheet
Use Part 1 to sharpen your pitch. Use Part 2 during practice presentations to give structured feedback.
Team Name: _______________________
Date: _______________________
Part 1 — Pitch Coaching Guide
The Judging Rubric
This is exactly how your pitch will be scored. Read it before you practise.
Criteria
3 — Strong
2 — Developing
1 — Weak
Problem Clarity
Real, specific, compelling
Somewhat clear
Vague or generic
AI Solution + ML Type
Named, justified, clearly explained
Named but vague
Missing or incorrect
Data & Bias
Sources + bias type + specific mitigation
Partial — mitigation missing
Not addressed
Ethics & Impact
Specific harm named + safeguards explained
Acknowledged but not addressed
Not addressed
Limitations & Failures
2+ specific failure scenarios
Vague acknowledgment
"It will always work"
Future Work
Ambitious, specific, grounded
Generic ("add features")
Not included
Pitch Quality
Clear, confident, strong visuals
Mostly clear, some rough spots
Hard to follow
Final score = 50% judge rubric + 50% audience popular vote at the exhibition.
Delivery: Do's and Don'ts
✅ DO
Open with a story, stat, or scenario that makes the problem feel real.
❌ DON'T
Open with "Hi we're Team X and today we'll be talking about..."
✅ DO
Use your slides as visual cues. Speak to the audience, not the screen.
❌ DON'T
Read bullet points off slides word for word.
✅ DO
Be honest about limitations and bias. It builds credibility.
❌ DON'T
Say "our AI will be 99% accurate" or "there are no ethical concerns."
✅ DO
Practice out loud, timed. Know your section cold.
❌ DON'T
Run over time. 7+ minutes signals you haven't prepared.
🎤 Section-Specific Coaching Tips
Slide 1 — The Problem (Hook them first 20 seconds)
Start with a story or stat that makes the problem feel real and urgent. Not: "Healthcare is important." Instead: "My grandmother forgets to take her medications every week, and nobody catches it until she gets sick." Make the audience feel it. This is your one chance to grab attention.
Slide 2 — The Solution (Show, don't tell)
Describe from the user's perspective: "She receives a reminder at 8am. She taps a button to confirm. The system tracks her compliance." Use a mockup, screenshot, or video. Saying "AI-powered medication tracking" means nothing; showing it means everything.
Slide 3 — How It Works (Name the ML type like it's obvious)
Don't say "we use machine learning." Say "we use a classification model trained on historical compliance data." Name it. Explain the training process in 2–3 sentences. Include a simple diagram showing: data in → model → prediction out. If you use an LLM, say which one and how (fine-tuned, prompted, RAG).
Slide 4 — Data & Bias (Specific = credible)
Name the bias type you're concerned about. "We might have sampling bias because our training data comes from a hospital in one city. We're addressing it by collecting data from clinics in 3 different regions." Vagueness loses points; specificity wins judges.
Slide 5 — Ethics, Limits & Failures (Own the problems)
Give 2 concrete failure scenarios: "If the system sends a reminder at 2am instead of 8am due to a timezone bug, the elderly user sleeps through it." "If the model was trained primarily on younger patients, it might misjudge adherence for older demographics." Honesty disarms skeptics and builds credibility.
Slide 6 — Future Work (Dream big, stay grounded)
Don't just say "add more features." Say: "In Phase 2, we'd integrate with pharmacies to auto-refill prescriptions when doses run low. In Phase 3, we'd build a predictive model to forecast compliance drop-offs before they happen." Ambitious but doable = impressive.
Handling Q&A
After your pitch, judges and classmates may ask questions. Strong answers start with: "That's a great point — here's how we've thought about it..." If you don't know: "We haven't fully solved that yet — but our current thinking is..." is far better than a defensive or vague response. Never lie or exaggerate. Judges respect honesty.
Part 2 — Self-Review Before You Present
Fill this in as a team before your practice pitch.
Opening hookWhat's our opening line or scenario? Would a stranger feel the problem immediately?
ML type & whyDo we clearly name and justify our ML type? Could someone explain it back to us?
Bias — named & mitigatedWhich specific bias type? What specific mitigation? Is it vague or concrete?
2 failure scenariosCan we describe two specific cases where our system produces a wrong or harmful output?
TimingHave we timed this out loud? Are we within 5–7 minutes?
Part 3 — Peer Feedback Sheet
Complete one per team you observe. Use specific, constructive language.
Rubric Scores
Circle or click a score for each criterion. 3 = Strong, 2 = Developing, 1 = Weak.
Problem Clarity
AI Solution + ML Type
Data & Bias
Ethics & Impact
Limitations & Failures
Future Work
Pitch Quality
⭐ What worked — be specific
🔍 One challenge — frame as a question
💡 One concrete suggestion
Part 4 — Post-Feedback Action Items
After receiving feedback from classmates and your teacher, list your top 3 improvements to make before the exhibition.