Классификация новостей
For AI News, we built an ML microservice for a startup working with a stream of news from multiple sources. The product aggregated news from various channels, analyzed content, and delivered a processed feed without duplicating materials.
Client
AI News
Period
2 weeks
Format
ML microservice
About the project
For AI News, we built an ML microservice for a startup working with a stream of news from multiple sources. The product aggregated news from various channels, analyzed the content, and delivered a processed feed without duplicating materials. When several news items with identical or very similar content entered the system, the task was not to show them as separate entries, but to merge them into a single consolidated item enriched with all available data. Within this pipeline, our module handled classification and helped structure the flow of materials inside the product.
The Challenge
We needed to build a service that automatically determines the topic of a news item and helps route the flow of materials within the news product. For the client, this was important as part of a broader content processing system: the startup needed not just to aggregate news, but to analyze it, eliminate duplication, and form a cleaner and more useful feed for the user.
Our Solution
- Built an ML microservice for automatic news classification
- Designed processing logic for incoming content streams from multiple sources
- Built a pipeline for determining news topics based on text content
- Delivered the service as a standalone module suitable for integration into a product pipeline
- Accounted for the system context of duplicate news handling and structured content delivery
- Prepared a foundation for scaling and further intelligent processing of news streams
Results
- ML news classification microservice delivered as part of a news stream processing system
- Module helps structure content within the product and simplifies further routing
- Solution integrated into the startup logic for deduplication and feed cleaning
- Users receive a cleaner, more coherent news feed instead of multiple near-identical publications
- Case demonstrates ability to develop applied ML modules as part of larger media products
Similar projects
Voice AI Trainer
For RZD, we developed an educational platform where employees could study learning topics and take knowledge assessments with an AI assistant via voice interaction — a modern alternative to conventional tests and formal training.
Анализ сентимента Telegram-каналов
A private client approached us with a hypothesis that Telegram channel publications may influence the behavior of some Moscow Exchange market participants. We built an ML module to analyze messages, detect company mentions, and determine publication sentiment.
Let's buildsomething extraordinary.
Ready to start your next project? Reach out and let's discuss how we can help you achieve your goals.