internXL Innovation Challenge

Hackathon Framework and Trainer Scripts

Welcome to the internXL Innovation Challenge! This program is designed to help students develop their ideas, build functional prototypes and MVPs rapidly, and master the art of pitching to a live audience.


Pre-Hackathon Training

Before the Hackathon, students will complete six one-hour training sessions designed to help them develop their ideas, build functional prototypes and MVPs rapidly, and master the art of pitching to a live audience.

  • Kickoff and Objectives (10 mins)
  • Training (30 mins)
  • Guided Exercise (15 mins)
  • Wrap and Homework (5 mins)

Training Sessions

Orientation

Course Introduction, Setup, Expectations

  • Course overview and objectives
  • Development environment setup
  • Key concepts and expectations

Session 1 - From Idea to Product

Transform broad ideas into buildable product definitions

  • Problem framing and user definition
  • Requirements & constraints identification
  • Business & scale considerations
  • Experience & usability planning
  • AI tools for PRD development (PRDKit, ChatGPT/Claude/Gemini)

Session 2 - Fundamentals of Building with AI

Make AI outputs reliable through effective prompting

  • Good prompt design principles
  • Controlling generation parameters
  • Testing & evaluation strategies
  • Choosing appropriate models (OpenRouter)

Session 3 - Data Sourcing, Cleaning & Evaluation

Ensure your AI is trustworthy through quality data

  • Sourcing public datasets
  • Data cleaning & preparation
  • Building evaluation sets
  • Quick sharing with Gradio/Hugging Face Spaces

Session 4 - Rapid Prototyping and Software Development

Build working prototypes in hours, not weeks

  • Rapid prototyping tools (Lovable, Bolt.new, Vercel v0)
  • Portability & scaling considerations
  • Demo-first mindset
  • Deployment options

Session 5 - AI Agents and Tool Use

Enable AI to act, not just respond

  • Understanding AI agents
  • Tool calling and schemas
  • When to use agents
  • Practical considerations (timeouts, retries, validation)

Innovation Challenge

Day 1 (Hackathon)

Coding sprints alternating with mentor talks

Time Block
9:00–9:20 Kickoff & rules (judging, submissions, logistics)
9:20–10:00 Mentor demo (inspiration; 30–40 min)
10:00–12:15 Coding Block #1 (2h15m)
12:15–12:45 Working Lunch
12:45–2:45 Coding Block #2 (2h)
2:45–3:15 Mentor demo
3:15–5:30 Coding Block #3 (2h15m)
5:30–6:00 Working Dinner
6:00–8:15 Coding Block #4 + Office Hours (2h15m; optional 15‑min mentor slots)
8:15–8:45 Mentor demo
8:45–9:00 Day 1 wrap (submission checklist, Day 2 brief)

Day 2 (Pitch Competition)

  • 3:00 PM — Submissions due (GitHub repo, demo link or 2‑min backup video, PRD from Session 1, data card & eval set from Session 3, 10‑slide deck)
  • 6:00–9:00 PM — Pitch ceremony

Pitch Format (per team)

  • 5 minute pitch (problem, solution, impact, product vision)
  • 3 minute live demo (show the happy path; backup recording allowed)
  • 2 min Q&A (judges ask clarifying questions)

Rubric

Pitch (40 points)

  1. Problem clarity & importance (0–10)
  2. Solution fit & differentiation (0–10)
  3. Impact within theme (0–10)
  4. Storytelling & delivery (0–10)

Demo/Technical (60 points)

  1. Working prototype depth (0–15)
  2. Quality of AI integration (0–15)
  3. Data handling & reliability (0–10)
  4. Scalability path (0–10)
  5. UX & usability (0–10)

Tie-breaker: Incorporation of mentor feedback between Day 1 and Day 2.


Guidance Handouts

For Students

  • Scope to one happy path. Cut features aggressively.
  • Repo first: create a private GitHub repo on Day 1, commit early/often, include .env.example and README setup steps.
  • Use AI tools: PRDKit for PRDs/flows, Lovable/Bolt.new/v0 for scaffolds, ChatGPT/Claude/Gemini for code snippets and iteration.
  • Document your data (data card + eval set). Use only public/synthetic datasets.
  • Record a 2‑minute fallback demo video.

For Mentors

  • Your role is to coach, not code. Guide teams to scope and prioritize.
  • Use demos to inspire: show your own AI projects, lessons learned, and practical tips.
  • During coding blocks, circulate and help teams unblock. Use office hours for deep dives.
  • Encourage teams to focus on clarity, reliability, and demo readiness over feature bloat.

Text 404-962-0908 with feedback and questions.

Copyright © Big Poppa Code & Progress and Fortune LLC 2025