Project Info

Frame

Inspiration

I was working at a big tech company this past summer on a research-adjacent team, and the method of storing information was mainly through google docs-style documents, where links would just be passed around via slack. This made it hard to find out what other people were working on, and moreso how that could relate to what I was working on.

What it does

It computes similarity scores for your documents, and represents the documents you have, the documents your peers have, as well as all the sources used for each document as a weighted graph and allows you to find other people's work and research that's related to yours.

How we built it

Used Distilbert for generating embeddings, Singlestore db with vector store, python + JS backend, with React for the frontend work. All of it is hosted on digitalocean droplets.

Challenges we ran into

I ran into problems working with the text data from the JS editor and providing it to the bert-endpoint.

Accomplishments we're proud of

I'm proud of the fact that it's a mostly working product, and that the similarity scoring actually works!

What we learned

Learned a lot about devops this time and about Singlestore's DB, I was fairly familiar with the other frameworks and technologies used.

What's next

I think it would be cool to keep building in more features, possibly ML-related.

Analysis

Compare with all teams

View

Metric

Abhiram Chennupati
12

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only

FlaskIn code
PythonIn code
ReactClaimed

2 of 3 appear in the indexed code. 1 claimed on Devpost could not be matched to code, which may simply mean the tool leaves no trace in the repository.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

12 KB

Source files

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

Feature verification

3 Verified4 Code-supported2 Claimed only1 Blocked

Document similarity scoring via vector embeddingsVerified
It computes similarity scores for your documents
Claimed on Devposthigh confidence
- data_helpers.py:90 — get_similar_papers fetches a paper's embedding and runs a SingleStore VECTOR similarity query (text <*> @query_vec) to rank other papers
- app.py:7 — /get_similar_papers endpoint wires the similarity function to a request
SingleStore vector database storageVerified
Singlestore db with vector store
Claimed on Devposthigh confidence
- data_helpers.py:1 — imports singlestoredb and connects via s2.connect
- data_helpers.py:106 — query casts embedding to VECTOR(768) and does a dot-product similarity search, confirming vector store usage
Tracking sources linked to papersVerified
all the sources used for each document
Claimed on Devposthigh confidence
- data_helpers.py:11 — add_source inserts into Sources and links via Papers_Sources join table
- data_helpers.py:162 — get_paper_sources joins Sources and Papers_Sources to list a paper's sources
Embedding generation for papers/sources (DistilBERT)Code-supported
Used Distilbert for generating embeddings
Claimed on Devpostlow confidence
- data_helpers.py:258 — get_embeddings posts text to an external HTTP endpoint (http://147.182.163.168:6969/embed) and returns the response; the endpoint itself is not in this repo so DistilBERT usage cannot be confirmed here, only that an embedding call exists
Python + JS backendCode-supported
python + JS backend
Claimed on Devpostlow confidence
- app.py:1 — the backend present in this repo is a Python Flask app; no JS backend code exists in this clone
REST API for adding/updating papers and sourcesCode-supported
implied by 'mostly working product' with similarity scoring
Claimed on Devpostmedium confidence
- app.py:55 — api_add_paper calls add_paper, but add_paper is never imported into app.py (only get_similar_papers, get_source_papers, get_db_connection are imported), so this endpoint would raise NameError at runtime
- data_helpers.py:28 — add_source's INSERT statement is malformed (missing closing parenthesis and a column name for the 4th placeholder), indicating this path is untested/broken
Viewing peers' documents / documents by ownerCode-supported
find other people's work and research that's related to yours
Claimed on Devpostmedium confidence
- data_helpers.py:185 — get_user_papers returns papers filtered by owner, and get_similar_papers can be called on any paper id, but there is no code that combines ownership with similarity to explicitly surface 'peers' work related to yours'
React frontendClaimed only
React for the frontend work
Claimed on Devposthigh confidence
Weighted graph representing documents, peers, and sourcesClaimed only
represents the documents you have, the documents your peers have, as well as all the sources used for each document as a weighted graph
Claimed on Devposthigh confidence
DigitalOcean droplet hostingBlocked
All of it is hosted on digitalocean droplets
Claimed on Devpostlow confidence

An AI agent derived these features from the project’s Devpost page and readme, then searched the code for each one. Verified features are backed by cited code; claimed-only features had no supporting code, which is not by itself proof a feature is missing.