Seminal
The magazine and atlas of EU-funded researchA new publication from the web and AI team at Scott Charnley

Atlas / Ukraine

European Innovation Council · Running

Hillel Hillel: AI-Driven Autonomous Learning Platform

Coordinator
Limited Liability Company Edmake, Ukraine
Runs
1 August 2026 to 31 July 2028
Status
Running
EU funding
€500k
Total cost
—
Partners
1 organisations in 1 countries
Type
European Innovation Council
Grant agreement
101313025

What the project is doing

In the project's own words, from its summary on CORDIS.

Project: AI-Driven Autonomous Learning Platform: Scaling IT Education via Affective AIThe Problem The EU faces a critical shortage of 9.7 million IT specialists by 2030. Traditional online education (MOOCs) fails to bridge this gap due to a ""completion crisis"" (5-15% success rate) caused by a lack of emotional support. Simultaneously, scaling human mentorship is blocked by instructor burnout, as educators spend 60% of their time on administrative routine.

Solution & Deep Tech Innovation We are engineering the first ""AI-University without teachers""—an autonomous ecosystem where AI handles 90% of the teaching cycle (explaining, verifying, supporting). Unlike standard text-based ""wrappers"" (like Duolingo Max), our core innovation is Multimodal Fusion. The system synchronously processes 5 data streams (video, audio, text, behavioral telemetry, time) to recognize user emotions (stress, frustration) in real-time and adapt the curriculum accordingly.

Key Components1. AI-Tutor (Avatar): A ""digital twin"" offering personalized explanations and emotional validation.2. Embedded IDE: A coding environment with instant feedback (<200 ms latency), preventing cognitive gaps.3. AI-Interviewer: An automated module for soft-skills training and mock interviews. Measurable Objectives & Impact Our goal is to advance from TRL 4 (lab validation) to TRL 7 (operational prototype) over 24 months.

Efficiency: Accelerate learning speed by 40% (Time-to-Competency) and reduce dropout rates by 50%.Accuracy: Achieve a Burnout Prediction F1-Score ≥ 0.85, outperforming current behavioral models. Economic: Reduce mentor administrative workload by 60%, delivering a 300-500% ROI for B2B clients. Implementation Backed by a proprietary ""Data Moat"" of 100,000+ historical student logs, we will execute a 24-month roadmap. This culminates in a ""Living Lab"" pilot involving 1,600 users across Ukraine, Romania, and Spain to validate the technology in a real-world environment.""

Who is involved

OrganisationCityCountryRole
Limited Liability Company EdmakeOdessaUkraineCoordinator

Source: CORDIS, European Commission. Data of 5 Nov 2026, reused under the Commission's reuse policy. View on CORDIS · Grant DOI