Inspiration
The challenge of the initiative for applied artificial intelligence to democratize access to AI for every business motivated us to create OptiFlow.
What it does
OptiFlow assists clients and consultants in identifying processes that can be enhanced with AI and proposes individual AI solutions. OptiFlow is an easy-to-use assistant that can help every company, no matter how large it is. The only input it needs is a company URL.
Workflow
OptiFlow scrapes the company website to understand the business model of the company, what they offer to clients, and what makes them unique. With this information, OptiFlow recognizes business functions and departments in the company. After choosing one of them a list of processes (from the chosen business function) that can be optimized with AI gets created containing company-specific processes. After asking 5 company- and process-specific questions to further determine potential risks, costs, and benefits we map a database containing existing AI tools with our results to potentially already find existing AI tools, from which the company can directly benefit. As a result, OptiFlow generates a complete use case for an AI solution for the company.
How we built it
We built OptiFlow using a React Typescript Frontend as well as a Python Flask backend server.
Challenges we ran into
Prompt engineering for good text summarization and understanding.
Accomplishments that we're proud of
Our application isn't fully dependent on gpt knowledge because we gather our information by scraping the website directly.
What we learned
More about prompt engineering, design and fast development
What's next for OptiFlow
Faster process, multithreading, and deployment
Log in or sign up for Devpost to join the conversation.