Seedlytics

Search Engine Home Page
Search Autofill
Streamlit Data Input
Streamlit Grapher
Search Engine Result Page, Business List
Search Engine Result Page, Business Info
Search Engine Result Page, Business Info 2

Inspiration

As undergraduate students experienced with the engineering design process, we were very interested in the next steps after prototype development—commercialization and bringing products to market. When developing our projects in high school, we ran into multiple issues regarding procuring prototyping materials as well as the software and technology to implement our ideas. For a new venture into any industry, developing a viable product and refining it will take money—lots of it. As such, early “angel” investors are an integral part of the “seeding” of startups which may eventually grow into the next Amazon or Airbnb. However, this critical stage is where most startups begin AND end their life cycles as many potential investors are rarely made aware of their presence or potential.

What it does

To make a marked change in this field, we have created Seedlytics, which offers potential investors a search-engine form of indexing and interacting with startup companies. Seedlytics will bridge the gap between avid investors and the crucial financial seeding stage by allowing users to search key sectors, geographical locations, prior funding, and names of companies (among other criteria) to encourage “matching” of interests.

How we built it

We used TypeScript, Flask, MongoDB (Atlas), Streamlit, and Python. We used a typescript home page that had a search bar. After searching, the backend would query the databases and return the companies with matching values. We used Flask to send requests to the backend, and Python to query the MongoDB Atlas database and clean data, then return a json object. The backend would then send the json object to the frontend, which would then display the matching companies. Streamlit was utilized for processing and visualization of financial data.

Challenges we ran into

We ran into issues with implementing and utilizing MongoDB Atlas at the beginning as we did not have prior experience with the package. About halfway into the hackathon, queries to MongoDB Atlas mysteriously stopped working, causing the backend to freeze. After much debugging, we attributed the cause to Flask not refreshing when MongoDB Atlas was connected.

We were also unable to fully optimize the sorting algorithm to work for all search filters.

Accomplishments that we're proud of

We are proud of being able to create an accurate search engine algorithm that can be tailored to the needs of investors and startups using advanced search filters. We're happy to have resolved problems integrating MongoDB Atlas into the web app. Additionally, we have been able to achieve great results with sorting through the implementation of MongoDB Atlas to vastly improve the search index speed.

What we learned

We learned that Streamlit is an excellent resource for data visualization and processing. It can vastly improve the visual aesthetics of the final product through built-in functionality. We also gained experience integrating frontend and backend frameworks.

What's next for Seedlytics

We will certainly need to incorporate a method of payment and/or a templated funding agreement between the startup and venture capitalists to facilitate the process of near-immediate funding (possibly through Streamlit).

Dataset: https://www.kaggle.com/datasets/yanmaksi/big-startup-secsees-fail-dataset-from-crunchbase