Circulumn: Data-Driven Academic Support for Women in STEM

Inspiration

Women in STEM often face invisible barriers: unsafe study environments, limited mentorship access, and financial stress that impacts academic performance. We wanted to build a platform that turns these challenges into measurable, data-backed solutions.

Circulumn empowers women to conquer college with safe spaces, meaningful mentorship, and real-time academic intelligence that are all powered by MongoDB Atlas.

What it does

Circulumn combines three core systems: 1) Safe Study Finder Students browse, filter, and review study spaces using tags like quiet, late-night, well-lit, and inclusive. Community reviews create transparent safety and confidence rankings. 2) Mentorship Matching Students submit a mentee profile (major, courses, interests, availability). Our backend computes compatibility using weighted overlap scoring: $$ Score = (CourseOverlap x 3) + (MajorMatch x2) + (InterestMatch) + (AvailabilityOverlap) $$ This prioritizes meaningful academic alignment and real support. 3) Analytics Dashboard (Live MongoDB Aggregations) All insights are computed directly from MongoDB Atlas: Hardest Courses → Average stress by course Stress Trends → Time-based grouping Top Study Spaces → Composite safety + rating score Safety Flags → Tag frequency detection No static data. No mock analytics. Everything is aggregation-driven and live.

How we built it

Circulumn was built as a full-stack web application using:

  • Frontend: HTML + Tailwind CSS for a clean, women-centered, glassmorphism UI
  • Backend: Node.js + Express REST API
  • Database: MongoDB Atlas

We structured the platform around multiple MongoDB collections:

  • User
  • MentorProfile
  • StudySpace
  • Review
  • StressReport

Instead of static data, we used MongoDB aggregation pipelines to power live analytics such as:

  • Average stress per course
  • Study space composite rankings
  • Safety tag frequency detection
  • Mentor compatibility scoring

Every dashboard metric is computed dynamically from Atlas, making the database the intelligence engine behind the platform.

Challenges we ran into

  • Turning data into insight: Storing data was easy but designing meaningful aggregation logic was harder. We iterated on scoring formulas to ensure analytics reflected real academic patterns.
  • Balancing scope in 24 hours: We wanted strong UI, deep MongoDB usage, mentorship logic, and analytics. Prioritizing core data pipelines helped us stay focused.
  • Making it truly women-centric: We avoided surface-level design choices and instead centered structural needs: safety transparency, mentorship alignment, and financial stress visibility.

Accomplishments that we're proud of

  • Building a fully functional multi-collection MongoDB architecture
  • Implementing real aggregation-driven analytics instead of placeholder charts
  • Creating a weighted mentor compatibility scoring system
  • Designing a safety-aware study finder powered by community input
  • Integrating financial wellness as part of academic equity

What we learned

  • MongoDB Atlas can function as both a storage system and an analytics engine.
  • Aggregation pipelines are powerful enough to drive full dashboards without third-party tools.
  • Academic stress, safety, and financial pressure are deeply interconnected variables.
  • Designing for women requires addressing systemic friction, not just adding aesthetic branding.

What's next for Circulumn

  • Integrating MongoDB Atlas Search for more advanced filtering and fuzzy matching
  • Adding scholarship and internship financial tracking
  • Building predictive stress modeling using historical trend data
  • Partnering with universities to deploy campus-wide
  • Rendering a live link for the general public to use Circulumn could evolve into a data-driven academic equity platform used across institutions to improve outcomes for women in STEM.
Share this project:

Updates