H+12026 / 2027
The first proof surface ships.
The first consumer surface for scent-memory translation ships through the Demeter rail. The loop begins with human context, candidate comparison, physical artifact, and response.
Roadmap / Proof before infrastructure
From first proof to the sensory-affective computation layer.
Scent-memory coupling is the wedge because scent can make lived experience physical enough to test, refine, and learn from. Scent is where Scentient starts. It is not where Scentient ends. Each horizon compounds the loop from signal to state to graph to artifact to feedback to updated state.
Demeter Fragrance Library archive
A named scent vocabulary, commercial history, and physical fulfillment rail that no model starts with.
Capture surfaces
Early interfaces for memory, atmosphere, spatial edits, comparison, and response.
First physical proof
Scent artifacts are tangible, shippable, and precise enough to create real feedback.
Learning loop
Selection, rejection, revision, revisit, and later response update the graph.
H+32028 / 2029
The sensory-affective layer opens.
The context profile, translation surface, and sensory direction layer open to selected developers and design partners. First integrations land where memory, atmosphere, and sensory response matter.
H+52030 / 2031
The substrate hardens.
The affective memory graph becomes more durable as memory, atmosphere, artifact, response, and later behavior compound into longitudinal signal. Scentient's advantage is no longer a single output; it is a learning substrate with an affective value function for memory-conditioned sensory direction. The category becomes legible beyond fragrance.
H+102035 / 2036
Memory-conditioned sensory infrastructure.
Olfaction was the wedge. The graph can extend to adjacent sensory surfaces: atmosphere, spatial context, ambient sound, directional light, haptic vocabulary, and thermal cues. Scentient becomes the sensory-affective direction layer for interfaces that need human context.
H+62031 +
Human context becomes the scarce layer.
As general-purpose intelligence becomes more capable, durable advantage shifts toward proprietary context: what people remember, choose, reject, revisit, identify with, and respond to over time. Scentient is built to compound that signal before the category is obvious.
The Ask / For design partners
Own the sensory-affective computation layer before the category names itself.
Why now
The window is small.
The window is not fifty years out. Models are becoming more capable now, which means the scarce layer shifts from raw model access to the texture of lived experience. Demeter Fragrance Library gives Scentient a living vocabulary of named scents and customer behavior. Scentient turns that advantage into a physical response loop before the category is obvious.
Why us
The graph is the moat.
Scentient has unusual access to three things at once: Demeter's long-running scent vocabulary, the physical fulfillment rail to test artifacts in the real world, and the computation layer that turns memory, atmosphere, artifact, and response into an affective memory graph. Closed-loop affective learning is the compounding advantage.
What we are building
The sensory-affective computation layer.
A machine-readable substrate for scent-memory first, then adjacent sensory surfaces: memory-to-signal translation, physical artifact generation, response capture, an affective memory graph, and an affective value function that improves through opt-in feedback.
The direction: as intelligence scales, the scarce layer becomes structured, consented affective context.