Team
Companion Chatbot
This team is at maximum capacity.
Project Concept
We want to get as close as possible to replicating the AI companion from the movie Her (2012)! The most amazing feature of the AI in that movie was how interacting with her felt very close to interacting with a real human being. The key things we will attempt to agentically achieve are as follows:
- Needs to maintain an absolutely coherent back story of its origin and its past and be able to use its memory to keep the conversation intriguing. We will likely use some sort of vector store in combination with agents that sift through the store to achieve this.
- Far more opinionated and less prone to “people please” than conventional LLMs. This means that it needs to check the coherency of the conversation at every step so that it can judge whether the user or itself is being conversationally coherent. For example, if you tell the bot something like “The sky just turned green!!!!”, the bot should respond with some sort of silly response like “Maybe you were looking at your lawn?”.
On the product side, we will just need to create a texting interface where the bot doesn’t have to respond synchronously to every text. This is to simulate realism of texting with a real-life friend.
Entry
Status: Submitted
Last saved: September 30 at 5:31 PM EDT
Team Roster (team is at max capacity)
Message board not available for this team yet.
Tianren Wang Team Lead RSVP Approved
Software Engineer at Qualifacts
Project ideation, team management, research
Software engineer with passionate interest in the intersection of neuroscience and AI. My primary focus on any tech events is to make friends with similar interests and values.
I am really interested in researching techniques that allow RL agents to think more like human. The area I am currently researching as a hobby project is grid and place cells and its application in deep learning. I am also interested in revisiting human-like chatbots using existing popular AI technologies.
I am researching a technique that allows a model to self-classify its own latent states. I was inspired to research this from deep learning work done on grid cells, which are believed to be the neural substrate that allows spatial (and potentially abstract) reasoning in the brain. If we understand the relationship between spatial and abstract reasoning, we can apply generally apply it to endless number of applications that require deep, strategical reasoning.
Sonal Bihani RSVP Approved
Data Scientist at Clio
memory agent implementation
I’m a Data Scientist at Clio, with previous experience at Uber, where I worked as a Data Analyst for 2 years. I’ve completed a Masters in Data Science from University of Waterloo
Machine Learning, Causal Inference and AI agents
I’m working on a establishing a causal inference setup at my company
Bhavnoor Kaur RSVP Approved
Software Engineer at Intuit
Designing and implementing guardrail agent
Bhavnoor Kaur is a Software Engineer at Intuit, focusing on AI. She is a Software Engineering student at the University of Alberta and is currently developing a GenAI experience for customer support.
Currently working on a GenAI powered experience to help support experts at Intuit assist customers.
Aarush Khurana RSVP Approved
Software Engineer at Qualifacts
Database setup and implementation
I’m an AI Engineer focused on shipping healthcare-grade AI features and integrations. I build full-stack systems—connecting React/TypeScript front ends to Node/Python services—deploying on AWS (S3/Lambda), enabling realtime with Socket.IO, and implementing vector DB/RAG pipelines. I validate reliability, security, and compliance for clinical workflows.
I enjoy chess, solving rubiks cubes, building legos, and sushi. I really love sushi I even have a membership at spoon and fork.
On a more technical note I enjoy working with LLM's to build data driven tools that can make a positive impact on peoples lives
The main project I am working on right now outside of work is an AI tool called EduReel. It is a platform that transforms students lives. Leveraging RAG, I created a platform where users can upload and ask questions about their notes, create flashcards, reels, quiz's, podcasts, etc.
Jacintha kurniawan RSVP Cancelled
Cofounder at Bridge ai knowledge
Did not attend
Builder at heart, loves to experiment and build stuff but mainly with focus on building stuff that people actually care about. I have an mba business background so translating features/products with real value and business roi is what i care about. I run an ai consulting and implementation where we build ai agent systems.
Looking to connect w engineers who are also keen on building something that provides real value
Ai smart project manager who manages everyone’s work and communication
Seyitan Oke RSVP Approved
Product Designer at Lululemon
UIUX design
Senior Product Designer with 7+ years shipping user-centered products across game development, fintech, and consumer mobile. Most recently at Unity Technologies, I led accessible gaming workflows that empowered developers worldwide to create inclusive experiences. My design approach bridges technical feasibility and user delight—whether building AR language learning prototypes, redesigning investment platforms at BMO Wealth, or scaling design systems for millions of users at Lululemon and PC Financial. I currently teach UX Design at Humber Polytechnic while exploring the intersection of spatial computing and everyday tools.
AI-augmented interfaces and agentic workflows, interactive event technologies and spatial computing (AR/VR), creator/developer tools, personal finance innovation, small business solutions
RingRing.com - Interactive event technologies for memory sharing
HoneyPot - An AI driven personal finance manager that's more reliable than Plaid
Masooma Suleman RSVP Approved
AI Scientist at CIBC
Vector store setup
I'm an AI Scientist at CIBC, where I work on applying Generative AI to build solutions, like automating the creation of complex reports using large language models (LLMs). My goal is to make these systems not just powerful, but also practical, secure, and easy to understand in the context of financial services.
Most of my day involves working with tools like Python and Azure to design, build and deploy AI applications that are both scalable and reliable. I'm especially interested in explainable AI, helping teams and stakeholders trust and understand what these models are doing, and why.
ML, GenAI, Agentic AI
ObservableAI