The Depth Lab

The idea

Devices that read what is going on in your head already exist. An EEG headband will tell you whether you were calm during twenty minutes of meditation. Devices that try to change it exist too: light flickering at chosen frequencies, sound, magnetic fields, vibration. Both kinds are cheap and easy to buy.

Nobody has built one that does both. Picture it. You sit down. It reads where you actually are, not where a fixed programme assumes you are, and starts with sound. Your breathing slows, so it shifts the light. Something in you tightens, so it backs off and finds another route. Tomorrow it begins where you got to today.

This is three weeks of people trying to build one.

Who is in the house

Embedded and analog engineers, signal processing and ML people, control systems engineers, XR developers, sound designers, industrial designers who make things people wear.

And the ones who will assume this is not addressed to them. Meditation teachers, anaesthetists and clinicians who move people between states for a living. Sleep researchers, because the threshold of sleep is one of the doors this tradition works with. Psychoacoustics and haptics specialists. Artists who work with light: the Dreamachine came from a painter and a mathematician, not from a lab.

Two conditions, both compulsory. You build, and you have a practice of your own. Most people who apply meet one.

The application asks for a specific device you intend to build, and one honest sentence about why. It does not ask for a meditation résumé.

Most people build alone. Teams form when they want to and come apart without ceremony when they stop being useful. Nobody is assigned to anybody.

The three weeks

A weekend hackathon gives you one attempt at an idea. Twenty-one days gives you fifteen.

Week one gets something crude running end to end. By day five you should have the roughest possible version working: a sensor feeding in, an output coming out, hard-coded, no enclosure, no machine learning, no polish. Integration is what kills projects in this field. Leave it until week three and week three is when you discover the pieces do not talk to each other.

Week two is building, testing, breaking and rebuilding. Mid-week each person walks somebody else through their build, unfinished as it is, and people from the wider village come in to try things.

Week three is about reliability. Does it work on someone who has never tried it? On a bad day? Twice in a row? Almost nothing in this field fails at the demo. It fails on the second person, and week three is where you find that out in private. Demo night closes the three weeks.

A day

The house sits together at 07:30 for an hour, before anybody opens a laptop. Breath, sound, mantra, stillness, rotating through the week. Five plain minutes on how the technique works, then the room does it.

Nine to five is the work. No standups, no check-ins, nothing scheduled. The village outside runs forty to eighty sessions a day and the house ignores nearly all of them.

At 17:30 the room sits again. Two or three evenings a week that hour becomes the test hour: somebody straps into another resident’s build, the room watches the traces, and the person wearing it says what actually happened.

By 22:30 it is quiet. Sleep is protected, because tired people write bad firmware and produce unreadable data. Sundays start in silence and then belong to you. Nobody takes anything for the twenty-one days.

Two committed hours a day. The rest is yours.

Testing on each other

Every prototype meets a stranger’s nervous system within hours of existing. That is the point of doing this in a house.

Test a device on yourself and you confirm your own hypothesis. You know what it is meant to do, so you feel what it is meant to do. Hand it to somebody else and they will tell you it did nothing.

Testing happens three ways, all opt-in. Informally, on the couch and on the beach. Quietly during the sits, for anyone whose build needs state data. And formally in the test hour.

Every stimulation device carries a physical kill switch, held by the person wearing it. Not a suggestion. A build requirement, and the reason people are willing to put each other’s work on their heads.

Goa, in October

The monsoon has just left. The rivers are full, the paddy is green, and the coast is emptier than it will be for the rest of the season. The house is on Mandrem beach in Pernem, the quietest stretch of Goa’s northern coast.

Some sits leave the house. A sunrise sit in the sixth-century Arvalem caves, cut into rock, cold and resonant. An evening on the ramparts of Tiracol fort. A night paddle through bioluminescent water on the darkest night of the month. Sensors welcome. No build day is lost to any of it, because the field session is that day’s sit.

The same short practice recorded in a cave, a boat and the house turns into a small study of its own. Does the place change the state?

The residency opens on the first night of Navratri. On Dussehra, the tradition’s own day for blessing your tools, the house walks to the temple and the instruments go on the altar. Sharad Purnima falls on 26 October, and that night the house takes the instruments down to the beach.

Across the three weeks about a thousand builders, researchers and founders pass through Edge City India, three hundred or so at any one time. They are the test population and the demo audience.

Practicalities

Sort your components before you come. Buy them wherever is easiest and carry them, or order from Indian suppliers and have them delivered ahead of you. I will send a list. Do not post parts into India from abroad, because somebody at this end has to clear customs on them. You fill in a short medical screening before you come. There is a safety officer who is not building, a 24-hour clinic in Mandrem, and hospitals forty-five minutes away.

Residents own everything they build and everything they measure.

Why I am running it

Because I could not find anybody else who was.

I have spent two decades manufacturing things people wear every day. Fine jewellery at volume, which means CAD, casting, tooling, workshops and supply chain. Wearable neurotech is arriving at the same form factors jewellery has always used, and at the same problem underneath them: making something a person will keep wearing. I have had a practice since school, been going on retreats since I was seventeen, and published translations of the source texts.

I am building one of these devices myself. Light driven at frequencies that shift what the brain is doing, closed through a Cerelog EEG board that reads the state and steers the light against it, tuned to a person’s own alpha and not to a fixed programme. Lucia N°03, Roxiva, Pandora Star and the Dreamachine all run predetermined sequences. The same light for every brain. Nobody has properly tested what a personalised closed loop can reach. That ceiling is unknown, and finding it is the work.