Sahha has shipped real-time scoring and on-demand biomarkers across its platform. Health scores and biomarkers now generate as user data arrives — eliminating the intra-day processing delays that previously separated data collection from computed insights.
This release also introduces two new SDK methods — getStats and getSamples — that let developers access health and fitness data directly from the user’s device, enabling instant display without waiting for server-side processing.
From batch processing to real-time delivery
Sahha’s previous scoring pipeline operated on batch cycles. Data uploaded from user devices was queued, processed in scheduled batches, and delivered as computed scores and biomarkers after a delay. For many use cases this was acceptable, but it created a gap between what users were doing and what apps could show them.
The new pipeline processes data continuously. As health data flows in from the SDK — sleep events, activity samples, heart rate readings, step counts — scores and biomarkers are computed and made available through the API and delivered via webhooks in near real time.
For developers, this means the Sahha API reflects the user’s current health state, not yesterday’s. For end users, it means the health intelligence in their app responds to their actions as they happen.
On-device data access: getStats and getSamples
Alongside server-side real-time scoring, this release introduces two new methods in the Sahha SDK that give developers direct access to health data on the user’s device.
getStats
getStats returns a daily summary of health metrics for any 24-hour period. Total steps taken, average heart rate, total minutes slept, active calories burned — the aggregated figures that apps commonly display on home screens and daily summaries.
This data comes directly from Apple Health or Health Connect on the device, formatted for immediate display. No server round-trip, no processing delay.
getSamples
getSamples returns an array of individual data samples within a specified time window. Step samples showing when a user started and stopped walking. Heart rate samples captured during a workout. Sleep stage samples for the previous night.
This granular data enables features like workout timeline views, sleep stage charts, and intraday activity graphs — rendered from local device data with zero latency.
The two-layer pattern
Together, these methods create a two-layer data architecture for health apps:
Device layer — immediate, local data for display and basic features via getStats and getSamples. Available the moment data is recorded on the device.
Cloud layer — computed scores, biomarkers, trends, and behavioral intelligence via the Sahha API. Available in near real-time as the server-side pipeline processes incoming data.
This pattern lets developers show users something immediately (today’s steps, last night’s sleep stages) while the richer computed intelligence (health scores, multi-day trends, behavioral archetypes) follows shortly after.
What 100+ biomarkers in real time enables
Sahha now delivers over 100 biomarkers across physical, mental, and recovery dimensions. With real-time computation, the full biomarker set becomes useful for time-sensitive features:
Responsive coaching. A workout completion triggers an immediate readiness score update. The app can adjust tomorrow’s training recommendation before the user puts their phone down.
Live health dashboards. Health metrics that update throughout the day as new data arrives, rather than refreshing once in the morning with yesterday’s numbers.
Timely interventions. A decline in sleep quality detected from last night’s data is surfaced in the morning — not 12 hours later when the insight is stale.
Event-driven workflows. Webhooks deliver score updates as they’re computed, enabling server-side logic that responds to health state changes in near real time — push notifications, content updates, coaching adjustments.
For developers already on Sahha
If you’re already integrated with the Sahha SDK and API, real-time scores and biomarkers are available now in both Sandbox and Production. Score delivery through existing API endpoints and webhooks is faster — no code changes required to benefit from reduced latency.
The new getStats and getSamples SDK methods require updating to the latest SDK version. Documentation covers the available data types, time window parameters, and response formats for both methods.
For new developers, the real-time pipeline is the default — there’s nothing to opt into. Start building at app.sahha.ai.