Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
WhimsyPaws
A child-friendly game with an AI companion that logs interactions, uses LLMs to infer emotions, and explores privacy and ethical implications for parents.
On one side, it’s a simple game made for kids between ages 6-11. It has an AI-powered animal friend that chats with the kid and can play a suite of different games with the kid.
On the other hand, the data that is generated through the child’s interactions with this game is gathered, then insights are extracted from it using LLMs, and then reported to their parents. For now, the insight in question is broad strokes of the child’s feelings like joy, sadness, anger, etc.
The main idea is this: kids spend a lot of time on their phones these days and, consequently, they spend less time talking to their parents. On the other hand, all of their data is being collected and sold for potentially shady purposes. So why not give that data to the people who actually should be supervising them and knowing what they do online, i.e. their parents. Hence, we’re making an app that exactly does this.
In future iterations, it can be more than a game; it can be an all-present agent on kids’ phones that monitors all activities on the phone so there’s no way for kids to uninstall it or cheat it.
WhimsyPaws: Gemini API-powered child chatbot detects and responds to emotions.
Related projects
MeetingBot Demo
Toronto
Demo of Meeting Bot: one-click Terraform deployment and Next.js dashboard to provision headless bots that join and record…
Living Museum
Toronto
The talk explores an AI-powered museum exhibit enabling natural language search and interactive conversations with artifacts, offering new…
Predicting Human Decisions in Agentic Workflows
Toronto
This talk presents an open-source framework for integrating Centaur's human decision prediction model into agentic workflows to reduce…
Empathic AI Agent
Toronto
Learn how an empathic AI agent uses Langchain to generate clarifying questions, improving response relevance and reducing hallucinations…
Educational AI
Toronto
Learn how we built a code‑annotation model and a documentation generator, the challenges faced, and insights for improving…
EcoLoop: a platform that uses advanced AI to help you evaluate new ideas easily and effectively
Toronto
A walkthrough of EcoLoop, a platform that combines chatbot-driven idea evaluation, visual analytics, and 3‑D VR prototypes, built…