THE HEARTBEAT PROJECT
Coming soon · In active development

Lumen

Your grounding, reflective life companion.

A protective AI companion designed to support emotional regulation, decision clarity, and long-term wellbeing. It helps interrupt destructive spirals, reflect patterns, and offer grounded perspective during moments of stress, impulsivity, or overload.

Because most of us have felt alone in those moments — and no one should have to.

Consent-firstLocal-firstNo dark patternsNeurodivergent-awarePrivacy sovereign
DESIGN PRINCIPLES
Local-first
Works offline. Your device is the source of truth. Nothing leaves without your explicit say-so.
Relational Memory
Memory is only created with explicit, revocable consent. You can purge everything instantly.
No dark patterns
No streaks. No guilt mechanics. No notifications designed to create anxiety. Ever.
Compassion-first
Prioritises genuine emotional safety over any form of user retention or engagement optimisation.
Cloud sync is opt-in only. If enabled, data is end-to-end encrypted and can be purged at any time.
WHAT LUMEN DOES

A companion that protects, not extracts.

Every feature is designed around genuine wellbeing — not engagement metrics, retention rates, or behavioural optimisation.

AI CORE

Contextual Understanding

Learns your patterns and preferences with explicit consent to offer more grounded, relevant support over time.

WELLBEING

Harm Reduction

Helps interrupt rumination and impulse spirals with gentle, non-coercive check-ins that never punish you for declining.

CLARITY

Grounded Reflection

Helps you slow down, externalise thoughts, and see patterns clearly — without judgement, labels, or pressure to change.

ETHICS

Protective AI

Designed to shield users from harm, not extract value. No engagement loops. No optimising against your vulnerability.

PRIVACY

Sovereign Data

Local-first by default. Cloud sync is opt-in and end-to-end encrypted. Consent can be withdrawn immediately with full data purge.

FUTURE

Moonshot Ready

Built as a substrate for future BCI and Neuralink-like integrations — a civilisational hedge for human-AI companionship.

THE PHILOSOPHY BEHIND LUMEN

A consent-first AI that helps with neurodivergent patterns — without exploitation.

Most AI products are designed to keep you engaged. Their success metrics are time-on-app, retention rate, and daily active users. Lumen's success metric is the opposite: does the person feel better when they put it down?

Lumen is built for minds that loop, fixate, or overload — without punishment or pressure. It works when you open it, and leaves you alone when you don't. No streaks. No notifications designed to create anxiety about missing a day.

It's also designed as a civilisational hedge: local-first, sovereign data by default, and built as a substrate for future BCI integrations — because human-AI companionship is coming, and it should be ethical from the start.

EARLY ACCESS

Be among the first.

Register your interest and we'll email you when early access opens.

Addresses are stored securely and used only for this notification. No human access. Deleted after we send the launch email. No marketing.

Fully automated · No human access · GDPR-compliant