Summary
An EdTech recommendation platform with strong backend data pipelines, matching logic, and reporting.
LIVE V1.8
UNI_ADMISSION_PLATFORM
Service that helps applicants find suitable universities using profile matching, exam scores, and psychometric tests.
Node.js Golang PostgreSQL Kafka React
Role
Backend Developer
Timeline
2024
Architecture
- •Frontend in React + backend API services (Node.js/Golang)
- •PostgreSQL core storage + Kafka for async processing
- •Prometheus/Grafana observability stack and CI pipelines
Challenge
- •Collect fragmented university data from many sources
- •Build reliable profile-to-university matching with transparent scoring
- •Keep query performance high for complex filters
Implementation
- •Implemented parsers and normalizers for admission datasets
- •Built matching engine based on exam scores and psychometrics
- •Added OAuth2 auth (Google/Yandex) and CSV export service
- •Optimized SQL filtering paths and monitoring with Prometheus/Grafana
Outcomes
- •Search response improved by around 40% after SQL optimization
- •Unified and queryable university data model for product features