Our Research Areas

We investigate how private, user-controlled intelligence can power a next-generation of service platforms. We explore technologies and models that protect user autonomy and enable meaningful personalization without surveillance.

Our work spans four areas that together explore a privacy-preserving, user friendly and user-owned ecosystem:

1. Consent and Control Layer

We are exploring how a social service booking platform can offer meaningful, transparent and user-driven control over personal information. With the aim to understand how such framework might support ethical data ecosystems and privacy-preserving computation.

Key themes:

  • how consent can be structured as granular, revocable tokens

  • which types of interactions should require explicit vs implicit consent

  • how local-first architectures reduce unnecessary data exposure

  • how users can verify, audit and reset their own data flows

  • what a “user-centered consent interface” could look like

2. On-device Preference AI

Investigating lightweight local AI models for private personalization. Researching whether user preferences can be learned directly on the device, without centralized data storing to keep user experience high. The goal to get a understanding if on-device AI can enable private, safe and scalable personalization for everyday digital services.

Key themes:

  • the size of a preference model to keep effectiveness

  • meaningful on-device signals for personalization

  • how models can adapt over time without compromising privacy

  • how preference graphs can represent stable long-tern tendencies

  • if federated learning should be used

  • user-control of reset, refine and audit their local model


3. Interactive Components

Exploring how AI can adapt UI components in real-time based on human emotions, intentions and contextual cues without storing intrusive behavioral data.

Key themes:

  • How affective signals and lightweight on-device preferences can guide moment-to-moment UI updates.

  • How user intent can be expressed through structured prompts that generate components rather than textual output.

  • How adaptive interaction patterns can be aligned with user autonomy, privacy, and wellbeing.

  • How a component-level reasoning engine can replace conventional feed algorithms built on surveillance and centralized profiling.

4. Ethical Data Economy

Researching theoretical value-flow models where user-generated signals could support future economic participation without exposing privacy. Investigating alternatives to today’s attention-driven digital economy. Explores weather valuable preferences privacy-protected preference signals learned on-device can form the basiis of more equitable value flows between users, services and platforms.

Focus areas include theoretical models for value attribution, privacy-preserving exchange mechanisms and conditions under which user data can support future economic participation without surveillance or behavioral manipulation.

Partnership Network

Fixmeapp collaborates with researchers, engineers and governance specialists across several regions. Our network includes partners and advisors with backgrounds in data privacy, machine learning, digital governance and emerging consent technologies.

We aim to bridge academia and industry through practical projects that explore the future of ethical digital infrastructure.

Contact us.

Partnership & Research Inquiries

Inquiries