The apps model for extended reality (XR) content is not going to work. No download, everything contextual. Driven by GenUI, enhanced by AI agents and most importantly developed from an open source library you can expand limitlessly. - Omnia plan for 2025 onwards.

Most smart glasses demos show someone reading instructions off a screen. That's not what we built. At Comer Industries (Rockford IL), we analyzed 1,025 measurement records across 13 assembly stations. The Settle Bearing step alone was failing 85.7% of the time. One unit needed 12 consecutive reruns on a single step with zero guidance on what to change between attempts. 76% of all failures had a measured value of exactly zero. The machine ran, captured nothing, logged a defect. Not a bad part. Not a broken tool. The worker initiated the cycle before a physical prerequisite was met — because nothing stopped them. Here's what we built: OmniaAgent connects Rokid smart glasses to the legacy OEM software running on the station. The machine completes a press cycle → glasses HUD advances automatically. Worker confirms a step by voice → station software advances in sync. On top of that: a CV model trained on footage from Comer's own equipment, their own parts, their own assembly line. Not a generic model. One that knows what a correctly seated bearing cup looks like on this fixture before this press descends. It checks before the cycle runs. It blocks the invalid cycle before the machine fires. The result: 71.9% of failures are directly addressable. $2.3M in projected annual savings — grounded in their actual production data, not estimates. One station. Path to 100. This is what enterprise AI on the factory floor actually looks like — agents integrated into legacy OEM software, CV models trained on the client's own equipment, error rates as the scoreboard. #SmartGlasses #ManufacturingAI #ComputerVision #EnterpriseAI #AIAgents #Industry40

9 Caveat builders should own: glass-grade latency still implies tethering or hybrid designs for the biggest variants today—but the representation you want on the device is exactly this flavor of predictive, dense, video-native model—not text-token physics.

8 The ViT-G scale shows the ceiling, but distilled ViT-B / ViT-L are the manifesto point: thermal and compute on the face are real. The research program is explicitly edge-tractable if you architect for distillation, not only cloud giants.

7 Action / dynamics understanding (e.g. on standard video benchmarks) supports hands-busy tasks: cooking, assembly, repair—where the camera shakes and the task is procedural. World model step-aware UI beats a raw LLM staring at JPEGs.

6 Strong depth from video (paper reports large gains on benchmarks like NYUv2) matters for AR occlusion, pinning, and plausibility—reducing the need to ship heavy depth hardware for every SKU (trade-offs remain; metric vs semantic).

5 On Ego4D, it pushes interaction anticipation—what object the user will engage with next. That’s the unlocked UX layer for GenUI: highlight the right thing before the hand moves, not after.

4 Unified visual memory (shared encoder for images video) is huge for glasses: the same representation for a still glance and continuous wear. That’s how you get coherent “where was that / what changed?” behavior without juggling two vision backends.

3 Deep self-supervision stacks the objective across encoder layers. Translation for product: you get usable signal throughout the stack, which matters when you distill or prune for on-device.

2 The dense predictive loss pushes the model to predict features for all tokens (masked and visible), not a patchwork where unmasked regions go blurry. For glasses, that means stable, localized perception under head motion—not smeared “video vibes.”

The @Hi_CYBERSIGHT Zenith is a strong prototype target for Omnia GenUI. It's a 39g binocular waveguide HUD with adjustable brightness from 10–1500 nits, IP54 protection, up to 8 hours battery, and crucially — it syncs with iPhone, Android, Apple Watch, and Garmin devices to pull real-time sensor data including power, heart rate, and pace via Bluetooth CYBERSIGHT. The display architecture is already built to accept a streaming data feed and render it in the FOV. What Omnia adds is the intelligence layer that decides what to render. The Zenith's phone-mediated connection to Apple Watch opens a clean data pipeline. Here's how Omnia taps it: 1. HealthKit as the data bus Common readable types include heart rate, HRV (heartRateVariabilitySDNN), SpO₂, and active workouts. Themomentum These are the exact signals Omnia's intent classifier needs. The iOS companion app requests granular HealthKit permissions per data type, then runs a persistent background query during an active workout session. 2. Near-real-time streaming via HKAnchoredObjectQuery To get data in near-real time, you declare an HKWorkoutSession to tell the Watch it's in active monitoring mode, then start an HKAnchoredObjectQuery — this triggers its updateHandler every time HealthKit receives a new sample Apple Developer, rather than polling on a fixed schedule. For heart rate specifically, this gives updates every few seconds during a workout. This feeds Omnia's intent classifier at a cadence appropriate for state transitions (low → moderate → high intent) without over-polling. 3. HRV as the deeper signal Beyond raw HR, Omnia queries HKQuantityTypeIdentifier.heartRateVariabilitySDNN — the RMSSD-derived HRV score Apple Watch exposes. This is what the BB-UAII policy engine uses to distinguish sustained high effort from momentary spike, which is the difference between showing a full directive or letting the display settle back to whisper mode. 4. Important constraint to be honest about You cannot access real-time heartbeat data directly; higher-frequency heart rate samples are only available within HealthKit for samples collected during Workouts and ECG sessions. Researchandcare Raw PPG is not exposed. This means Omnia's Apple Watch pipeline is latency-bounded by Apple's sensor-to-HealthKit write cycle — practically 3–5 seconds for HR, longer for HRV. That's fine for display density decisions (which operate on 5–15 second state windows) but rules out sub-second haptic/audio sync without a dedicated wristband like the Neural Band.

GenUI applied to smart glasses becomes much more than what Google research is currently discussing in the paper. UI that adapts and disappears when needed in such an obtrusive way in the user FOV is what can drive us towards true intent-driven interfaces.

"The recent proliferation of powerful LLMs has enabled the generation of UI code directly from unstructured natural language prompts. Our approach differs by tasking the LLM to generate entire, interactive, and data-driven web applications from a single prompt, effectively acting as an autonomous web developer."