Not every proof of what I build sits behind an NDA. Dreamalizing is an Elk Solutions product I can show in full: an AI that helps people explore their own dreams — privacy-first and bounded, with the marketing site live and the app in pre-launch.
It's not a client case. It's a studio case: I conceived, built, deployed and now run it myself. That's exactly why it shows, more sharply than any client project could, how I build controllable AI: AI that stays professional even with the most personal and sensitive input imaginable.
The challenge
Dream content is an extremely privacy-sensitive category. People share things in it they tell no one else. That imposes two hard requirements that seem to pull in opposite directions:
- AI has to be useful — otherwise it's just an empty text field
- AI must never pretend to know the truth — no therapy, no diagnosis, no "here's what your dream means"
A naive AI product picks one side: either an assertive dream-interpreter that fakes certainty, or a harmless journal with no value. Both are wrong. The real challenge was to use AI as a guide within strict safety and privacy boundaries — useful, yet honest about what it doesn't know.
The solution
I built Dreamalizing around a few deliberate choices, each one mirroring an Elk Solutions design principle:
- Guided exploration instead of black-box output. The AI doesn't interpret; it asks targeted questions. The user stays the authority over their own dream; the AI only helps explore.
- Bounded AI with clear limits. The model is explicitly barred from therapy, diagnosis or medical claims. Those limits live in the system, not in a disclaimer at the bottom.
- Privacy-first storage. Dreams are stored encrypted (AES-256-GCM, per-user key) on our own infrastructure — not at a large cloud vendor.
- Local inference. The language models run locally, enforced through configuration, so sensitive input doesn't leak to external AI services by default.
- Opt-in pattern recognition. Connections across multiple dreams are only sought when the user deliberately chooses it.
- Deletable and traceable. A deletion request actually wipes the data completely (GDPR), and the chain is auditable.
What this proves
Dreamalizing is a deliberately hard test case: if you can keep AI professional and bounded here, you can do it anywhere. It concretely shows that Elk Solutions:
- takes AI to the real world, not just a demo — the marketing site live and the app in pre-launch (iOS beta via TestFlight, Android coming), with a payment model and 14 languages already in place
- handles sensitive data with encryption, own infrastructure and local inference
- designs guardrails that are enforceable, not cosmetic
- can take an idea from concept to running product — including deployment, payments, maintenance and ongoing development
That last point isn't a detail. Most app ideas never reach a store at all: Apple rejects roughly one in four submissions, more than a third of all apps are abandoned within two years, and the stores purge those dead apps by the million. A product that makes it through App Review, TestFlight, payments and 14 languages — and keeps running — isn't a loose hobby project; it's proof of shipping capability.
The same principles — context as truth, measurability, governance and clear boundaries — are what I apply in client work around AI in production.
See it for yourself
Dreamalizing's marketing site is live; the app itself is in pre-launch (iOS beta via TestFlight, Android coming). You can explore the site, and if you want to put AI to work on sensitive or high-risk data, I'm happy to think through how to keep that bounded and controllable.