For connected-device engineering partners
Build the experiment that tells us whether DryNight should exist.
We are seeking an engineering partner for Stage 0, a bounded non-human bench feasibility program. The objective is not a polished wearable. It is a defensible answer on sensing feasibility, practical architecture, and what should happen next.
The assignment
From open sensing question to reviewable bench evidence.
Input
A defined clinical hypothesis, evidence map, risk register, and candidate modalities.
Work
Architecture, fixtures, electronics, prototypes, experiments, and documentation.
Output
An evidence-backed advance, redirect, or stop decision.
Stage 0 work packages
Six packages, one technical decision.
The sequence can be adjusted by the engineering lead, but every package must leave a reviewable artifact and an explicit gate.
-
A
Measurement architecture
Define candidate sensing modalities, placement assumptions, reference measurements, and the minimum signal that would justify further work.
Deliverables
- Architecture decision record
- Bench test matrix
- Risk-ranked modality shortlist
Exit gate
A testable measurement plan with explicit rejection criteria.
-
B
Coupling and body-interface bench model
Build repeatable fixtures that represent placement, contact pressure, coupling variation, motion, and posture before any on-body work.
Deliverables
- Fixture specification
- Coupling material comparison
- Repeatability protocol
Exit gate
A bench setup that can separate sensor behavior from fixture noise.
-
C
Low-power electronics and acquisition
Select the analog front end, processing path, power strategy, and data logging needed for Stage 0 experiments.
Deliverables
- Electronics block diagram
- Power and thermal budget
- Instrumented bench prototype
Exit gate
A data-acquisition path that captures raw signal and known test conditions reliably.
-
D
Wireless and private cue path
Prototype the minimum reliable link between sensing hardware, a haptic cue endpoint, and bench logging. No clinical alert logic is claimed.
Deliverables
- Link budget and protocol choice
- Haptic endpoint prototype
- Security and failure-state notes
Exit gate
A repeatable bench demonstration with observable delivery and failure states.
-
E
Prototype fabrication
Create fast, testable mechanical iterations for the sensor housing, coupling stack, and dock without locking a production form factor.
Deliverables
- Fabrication files
- Material and tolerance notes
- Bench-ready iterations
Exit gate
Hardware that supports the experiment without being presented as a finished product.
-
F
Bench data and methods package
Deliver a reviewable record of methods, raw data, known limitations, results, and the advance, redirect, or stop recommendation.
Deliverables
- Data dictionary and raw files
- Reproducible analysis
- Decision memo with open risks
Exit gate
An independent engineer can understand what was tested and why the conclusion follows.
Operating principles
Evidence has to travel with the hardware.
A prototype without its methods, raw data, and failure conditions is not a Stage 0 result.
- No human-subject testing in Stage 0
- No performance language from design renders or simulations
- Instrument the failure modes, not only the expected signal
- Preserve raw data and test conditions for independent review
- Choose the next experiment from evidence, not sunk cost
What DryNight has already defined.
Problem and claim boundary
A scoped pediatric hypothesis, prohibited claims, and current regulatory limits.
Evidence and risk file
Verified literature, competitor precedent, and the failure modes the experiments must address.
Founder access
Direct decisions, lived problem context, and rapid feedback on family experience.
Decision discipline
Permission to redirect or stop when the data does not support the next build.
Plan the Stage 0 bench program.
If your team can own connected sensing from measurement architecture through a reviewable bench data package, tell us how you would structure the work and where you expect the first failure.
Do not submit urgent medical questions or private information about a child or patient. Review the privacy notice before sending.