Inspiration
Marketing Research involving cultural references like MEMEs are often lost between the generational gaps. Because of this, market researchers look for ways to make this process more extensive and easier. However, most agentic agents, including ChatGPT, have a difficult time deciphering the cultural context of memes.
What it does
An agentic AI used for translating the cultural context of memes for industry purposes like marketing and intra-organizational communication.
How we built it
We implement a unique agentic multimodal workflow with Nemotron-Nano-4B-v1.1 as the primary driver, webscraping, and OCR models.
Challenges we ran into
Implementing the Nemotron in Google Colab.
Accomplishments that we're proud of
We're proud of our functional agentic multimodal workflow that utilizes Google Cloud's OCR model and NVIDIA's Nemotron models in combination with web scraping.
What we learned
We learned about how to implement an agentic AI pipeline.
What's next for Memotron
We plan to expand this beyond memes towards general culturally relevant events and expand the modality of media it can take in.
Built With
- css
- cuda
- flask
- google-cloud
- html
- javascript
- ngrok
- nvidia
- openai
- python
- react
- supabase
- supify
- typescript
- vscode

Log in or sign up for Devpost to join the conversation.