Case Studies

How Infonexs products run in real operations: our own platform and engineering, deployed for education and consumer app teams.

Client: AiEdvisor, a study-abroad and international education advisory service
Product: AIHub content and knowledge engine

Challenge: Admission policies and study-abroad information change often. Advisors wrote long-form guides and policy updates by hand, which limited how much content the team could publish each month.

Solution: Country policies and school information were organized into an AIHub vector knowledge base. The engine drafts in-depth articles and social updates, and an advisor approves each piece before it is published.

Result: Shorter production cycles for expert content, with every draft citing the policy sources it was based on.

Client: Almamind, a language training and education assessment platform
Product: AIHub adaptive question-generation module

Challenge: Fixed question banks offered little individual diagnosis, so teachers could not quickly create practice targeting each student’s grammar weak points.

Solution: A question-generation microservice built on semantic structure and a knowledge graph. It uses each student’s answer history to produce a grammar diagnosis and targeted practice questions.

Result: Personalized grammar diagnosis and practice sets generated on demand, with rule-based checks on grammar explanations.

Client: MohiGuide (毛孩街), a pet lifestyle platform
Product: Infonexs edge and cloud image-processing pipeline

Challenge: Users upload many pet and product photos with cluttered backgrounds. Sending every photo to the cloud for background removal added bandwidth cost and noticeable delay.

Solution: A hybrid design: background removal runs on the device first, with a Cloudflare-based cloud service handling the rest. This keeps latency low while the cloud side absorbs peak demand.

Result: A production mobile app that shows our team delivers cross-platform clients, on-device processing and cloud pipelines together.