How dgree.io Built a Course Marketplace That Aggregates Thousands of Learning Options Into Personalized Paths

The online learning market is enormous - and fragmented. Thousands of courses exist across Udemy, Coursera, LinkedIn Learning, specialized industry platforms, and independent providers. For someone looking to learn a new skill, the problem isn't finding a course. It's finding the right course among thousands.

dgree.io's founders saw an opportunity: a marketplace that aggregates courses from every major provider into one searchable platform. Users would search once, compare options side by side, and start learning - without visiting 15 different websites.

The technical challenge was significant. Each course provider structures their data differently. Some offer APIs. Others don't. Course metadata - title, description, price, duration, skill level, language - varies wildly in format and completeness. Normalizing data from dozens of sources into a consistent, searchable catalog was the core engineering problem.

Beyond aggregation, dgree.io wanted personalization. Not just "courses like this" recommendations, but actual learning paths - sequences of courses that build on each other to take a user from beginner to competent in a specific skill area.

dgree.io was a startup with a strong vision and limited technical resources. They needed a development partner who could take a concept - sketched on whiteboards and validated through user interviews - and turn it into a working platform.

easy.bi proposed a full product development partnership: from concept refinement and prototyping through to a production-ready marketplace. The team's experience with ElasticSearch for complex search applications, web scraping for data aggregation, and ebiPlatform for rapid backend development meant dgree.io could work with one partner instead of assembling specialists from multiple vendors.

The easy.bi team began by defining the concept alongside dgree.io's founders. User personas, core workflows, and technical architecture were established through collaborative workshops before any code was written.

Multi-source data aggregation. The team built a data ingestion pipeline that pulls course information from providers through two channels: API connections for providers that offer them, and web scraping for providers that don't. Each source feeds into a normalization layer that maps provider-specific data formats to dgree.io's common data model - ensuring consistent metadata regardless of origin.

ElasticSearch-powered discovery. The normalized course catalog is indexed in ElasticSearch, enabling sub-second search across thousands of courses. Users filter by topic, skill level, price, duration, language, and provider. Search relevance tuning ensures that the most applicable courses surface first - not just the most popular or most recent.

Personalized learning path engine. Beyond individual course search, the platform suggests learning paths: curated sequences of courses that progressively build a skill. The recommendation logic analyzes a user's stated goals, completed courses, and browsing behavior to suggest next steps. A user exploring Python programming sees a path from basics to data science to machine learning - with the best-rated course for each step pre-selected.

The frontend was built as a responsive web application using Angular and Vue for different platform components, with Ionic enabling a mobile-optimized experience. OAuth2 authentication through a custom ID service allowed users to save progress, bookmark courses, and maintain their learning history across devices.