Our Story
What Inspired Us
I’ve always believed that we live in one of the most pivotal times in history—where knowledge is more accessible than ever, yet misinformation spreads at an unprecedented rate. It’s frustrating to see false claims shaping public opinion while real breakthroughs in science and technology go unnoticed.
— Israel
During one of our brainstorming sessions, Ivan and I discussed this problem.
Ivan shared his perspective on misinformation and how AI could be a powerful tool for fact-checking. He also pointed out how mainstream media often prioritizes controversy over progress, making it easy to lose sight of humanity’s advancements. That conversation sparked an idea:
What if we built something that not only helped people uncover the truth but also reignited their sense of curiosity and belief in innovation?
That’s how Flow Innovation was born.
What We Learned
Building Flow Innovation was an eye-opening experience. We learned:
- The depth of misinformation – Fact-checking isn’t just about detecting falsehoods; it’s about ensuring credibility and trust. We explored AI techniques to make this process more efficient.
- How to integrate AI with real-world data – We had to think critically about how to use natural language processing and external fact-checking APIs to verify claims effectively.
- The challenge of engagement – People aren’t always searching for innovation news. We had to design a platform that makes discovery effortless and engaging.
- The balance between speed and accuracy – AI-driven fact-checking needs to be both fast and reliable, requiring optimizations in claim processing.
How We Built It
Tech Stack
- Frontend: React
- Backend: Flask
- Database: SQLite
- APIs: Google Fact Check API, Wolfram API, NewsAPI
- AI/ML: NLP (spaCy), TF-IDF + Cosine Similarity
Development Process
We started by outlining the core problem: misinformation is rampant, and innovation often goes unnoticed. From there, we designed two key features:
- AI-Powered Fact-Checking – We used NLP and Google’s Fact Check API to validate claims efficiently.
- Innovation News Feed – We integrated NewsAPI to curate the latest advancements in technology and science.
Once we had a solid foundation, we refined the UI/UX to create a smooth and intuitive experience. We kept the interface minimal, focusing on accessibility and readability.
Challenges We Faced
Every project has its hurdles, and Flow Innovation was no exception:
- Balancing scope and execution – Initially, we considered integrating goal tracking but had to evaluate whether it aligned with our core mission. Over time, we realized that fostering an innovation mindset naturally ties into goal-setting and progress tracking, making it a valuable addition.
- Optimizing fact-checking performance – We didn’t want every claim to hit an external API, so we implemented NLP-based similarity detection to reduce redundant checks.
- Ensuring a meaningful user experience – Presenting fact-checking results in a way that is clear and actionable was a challenge. We iterated on the UI to make insights digestible.
- Aligning with Hack for Humanity rules – Competing for multiple awards (1st Place, Best Use of AI/ML, Wolfram Award, and more) required strategic adjustments. Our backend engineer even completed a YouTube playlist on the Wolfram language to integrate it effectively. We also implemented additional machine learning features, even after having a fully functional app.
Despite these challenges, we stayed committed to our goal: making truth and innovation more accessible.
The Impact
We didn’t just want to build an app—we wanted to create something that informs, inspires, and empowers. Flow Innovation helps users:
- Uncover the truth with AI-driven fact-checking.
- Stay ahead by exploring cutting-edge innovations.
- Regain optimism by seeing how humanity is progressing.
This project is more than just a tool; it’s a statement that progress matters, and truth should always be within reach.
Built With
- flask
- flask-jwt-extended
- google-cloud
- google-fact-check-api
- javascript
- newsapi
- numpy
- pandas
- python
- react
- requests
- scikit-learn
- spacy
- sqlalchemy
- sqlite
- wolfram-technologies

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