7:45 AM
The Commute Brief
Before she parks the car, she knows more about her day than she used to know after thirty minutes at her desk.
Build NotesConcept
Focus Frames
A Second Brain, For Smarter Work
The AI-Powered Wearable for the Knowledge Worker
Bringing enterprise intelligence out of the chatbot
and into the physical world of work.
The problem
The enterprise has never had more data.
Revenue in Salesforce.
Cost structures in SAP.
Customer escalations in ServiceNow.
Workforce allocation in Workday.
And yet.
A VP of Sales hears a number in a meeting
and has no way to verify it in the room.
A CFO walks out of a board prep session and realizes
mid-hallway that something might be wrong.
A product manager loses critical information
to a hallway conversation no system ever captured.
The problem is not information access.
The problem is contextual information access:
the right data, at the right moment,
in the flow of how knowledge workers actually work.
The concept
Focus Frames
Enterprise glasses that function as a second brain for the knowledge worker.
Ambient Audio Intelligence
The glasses capture real-time audio during meetings, hallway conversations, and whiteboard sessions. AI structures the output: summaries, action items, decision logs, and verification flags.
Enterprise Data on Demand
Natural language voice queries against Salesforce, SAP, Workday, and ServiceNow. Ask a question, get a synthesized answer, no laptop, no context-switching.
Proactive Contextual Intelligence
In its most advanced mode, the system listens to ambient conversation, identifies where enterprise data adds value, and surfaces it without being asked.
The architecture
Four layers. One seamless experience.
- <3sVoice response time
- <15sComplex analysis delivery
- 9Horizon 1 use cases
- 4Architecture layers
A day with Focus Frames
A day in the life.
Sarah Chen. VP of Commercial Operations. Her workday, reimagined.
9:00 AM
The Pipeline Review
“Jake, can we double-click on that 2.8 number?”
This is the moment we designed the entire product around. Not catching someone being wrong. Having the data to ask the right question, in the room, when it matters.
10:30 AM
The Hallway Moment
“Capture that.”
A casual hallway conversation. An engineer mentions a critical API timeline change. Sarah touches the temple of her glasses. The last 90 seconds are saved, structured, and routed to the right team. Nothing fell through the cracks.
Half of the most important information in any company travels through hallways, not email.
12:30 PM
The Working Lunch
“Give me a comparison of our top ten accounts: revenue growth versus support ticket volume, last two quarters.”
The AI queried two systems, ran the correlation, and pushed the analysis to her email. Fifteen seconds. She had it before she finished her sandwich.
2:00 PM
Board Prep
“We are at 112. Benchmark median is 110, top quartile is 118.”
Nobody left the room. Nobody opened a laptop. Nobody sent a Slack message and waited forty-five minutes. The answer was there, in the conversation, when it mattered.
4:30 PM
End of Day
Six meetings. 4.5 hours. Every key decision captured. Every open item tracked. Two hallway conversations that would have disappeared are now in the system of record.
Nothing fell through the cracks. Nothing relies on memory alone.
A second brain, for smarter work.
How we think
Design thinking behind Focus Frames.
This is how Tenuto Labs approaches applied AI. We start with human friction, not technology.
- 01
We started with friction, not features.
Before designing any capability, we mapped where enterprise knowledge workers lose leverage during their day. The bottleneck was never data access. It was contextual data access: the right data, in the moment, without breaking flow.
- 02
15 use cases. Two horizons. One design principle.
We mapped every capability against two axes: where in the worker’s day it creates value, and what type of value it delivers. Nine use cases are buildable on existing hardware today. Six require future display APIs. Horizon 1 is useful without waiting for Horizon 2.
- 03
A roadmap built around trust, not features.
The hardest design decision was sequencing. Ambient proactive intelligence is the most valuable capability, but also the most behaviorally disruptive. We designed Horizon 1 around on-demand explicit queries. Trust is built before the system starts listening without being asked.
Use case matrix
- Pre-Meeting BriefH1
- Voice Data QueryH1
- Hallway CaptureH1
- Post-Meeting SynthesisH1
- On-Demand AnalysisH1
- Ambient CRM UpdateH1
- Cross-System CorrelationH1
- Whiteboard CaptureH1
- Customer Visit IntelligenceH1
- Proactive Ambient IntelligenceH2
- AR Data OverlayH2
- Live TranslationH2
- Negotiation SupportH2
- Onboarding AccelerationH2
- Emotional Intelligence LayerH2
System architecture
Layer 1Sensor LayerThe Glasses
- 5-mic audio array
- 12MP ultra-wide camera
- Open-ear speakers
No AI processing on-device. Glasses serve as sensor array and output device. All data streams to the companion mobile app.
Layer 2Edge ProcessingMobile App + Edge Cloud
- Speech-to-text (STT)
- Speaker diarization
- Edge AI inference (<3s)
Companion app performs initial processing. Edge cloud handles latency-sensitive AI inference for sub-3-second voice query responses.
Layer 3Enterprise IntegrationSecure API Gateway
- Salesforce / HubSpot
- SAP / NetSuite / Workday
- Custom enterprise APIs
Secure gateway translates natural language into API calls. Enterprise IT controls access, audit logging, and data residency compliance.
Layer 4AI Reasoning + SynthesisCloud AI Engine
- LLM synthesis (Claude / GPT / Llama)
- RAG with enterprise data
- Proactive ambient analysis
Core AI engine synthesizes multi-source data, generates contextual insights, and powers ambient intelligence. All outputs grounded in enterprise data via RAG.
Output channels
- Voice Response<3 seconds
- Push to Email<3 seconds
- Push to Slack/TeamsReal-time
- AR Display (H2)Real-time
Enterprise systems of record
- Salesforce
- SAP
- Workday
- ServiceNow
- HubSpot
- Jira
- Snowflake
- Custom APIs
- Layer 1Sensor Layer5-mic array, 12MP camera, open-ear speakers. No AI on-device.
- Layer 2Edge ProcessingSpeech-to-text, speaker diarization, edge AI inference <3s.
- Layer 3Enterprise IntegrationSecure API gateway to Salesforce, SAP, Workday, ServiceNow.
- Layer 4AI ReasoningLLM synthesis, RAG with enterprise data, proactive analysis.
Why now
The hardware is shipping. The SDK is open. The design thinking was missing.
Meta launched the Wearables Device Access Toolkit in late 2025. Ray-Ban Meta glasses have sold millions of units. Generative AI can synthesize data across enterprise systems in seconds. Every component exists. What has been missing is the design thinking to connect them for the enterprise knowledge worker. That is what we do.
Horizon 1
Buildable Now
Voice queries, pre-meeting briefs, hallway capture, post-meeting synthesis, on-demand data analysis, ambient CRM updating, cross-system analysis, whiteboard capture, customer visit intelligence.
9 use cases on existing Meta SDK hardware
Horizon 2
Future Capabilities
Proactive ambient intelligence, AR data overlays in-lens, live translation with enterprise terminology, negotiation support with real-time data surfacing, onboarding acceleration for new employees.
Requires future Meta display API access
Tenuto Labs
Applied AI Design Studio
We design applied AI systems that remove friction from critical decision loops. Focus Frames is one example of how we think.
Focus Frames is a concept prototype. Tenuto Labs designs the AI and software layer.