AI Productathon!

Theme:

“Build AI Products that Solve Real Problems”

---

Goals:

• Prepare team for future projects where we have to build products with AI Integration
• Understand the new possibilities that AI brings us
• Allow members to build and showcase new projects and make their own tech choices
• Incentivize research and personal growth
• Teambuilding & bonding

Tracks:

• Impact to Organization
• Productivity (Personal Productivity or Work productivity)
• AI for Impact (Create some positive impact in society any way you can with your product)
• Devtools
• AI for Learning
• Open Track

Team formation:

• 2 members with a synergizing skillset

Time period:

• Four weeks, demo starts from Week 2 for people that are ready
• Starts on Friday, April 25th after team finalization

Judging:

• Showcase & present your product at the end of the time period.
• The winning team gets goodies (tbd) and bragging rights 🥳
• Let's not get too swept away with ✨vibe coding✨, this is a learnathon as much as it is a hackathon

Criteria:

• Proper utilization of AI
• Real world impact & Practicality
• Technical implementation
• UX & Usability (We are known for our designs!)
• Solution to problem

Essential topics to look into:

• Interfacing with AI:

• Rest APIs
• SDKs like openai, Vercel ai SDK
• Running AI models locally through ollama, lmstudio, etc

• Prompt Engineering

• Essential prompt engineering techniques to achieve your goal
• Essential concepts like X-shot prompting, providing context, providing examples, etc.
• Setting up guard rails.

• AI Ecosystem:

• Popular LLM platforms like ChatGPT, Gemini, Claude, grok, etc
• Ollama, LMStudio Huggingface, etc
• Popular terminologies like Parameters, Context window, etc
• Frameworks (optional): langchain, huggingface transformers, etc

• UI/UX for AI

• The most common UI to interface with AI is chat interfaces
• Other notable ways are through mediums like voice, user action recording, mobile apps, browser extensions, VSCode plugins, embedable chatbots, providing public APIs for other people to build apps on top of, etc

What we are doing

• Building amazing products on top of existing solutions
• Learning enough to adapt ourselves to this new turn in the market

What we are not doing

• Being AI Researchers / Scientists. We are engineers, we build stuff. We solve real problems, for real people.