WHOOP AI Coach & Data Time Machines (Vol 6)
What WHOOP Coach teaches us about time series data and AI.
{Three, Curated, Perspectives} on generative AI and vector databases.
Finally, a fitness coach to answer what you really want to know, like “Should I have that second doughnut?” 🍩
WHOOP, “The Human Performance Company,” launched an AI assistant for its wearable biometric bracelet, WHOOP Coach. It uses your goals, biometric data, and performance science to answer health questions with AI. Their use of AI illustrates an app that demands time-series data — this week’s {The, Weekly, Vector} theme.
For more on the WHOOP app, read: Fitness training just got a lot more personal in Club Industry.
Why Your App Might Need a Data Time Machine
Apps like Whoop’s AI Coach benefit from the technology equivalent of a “Data Time Machine” to help answer questions like:
When was my recovery the worst in the last year?
How did my sleep last month compare to others my age?
Do I sleep better in the winter than in the summer?
By combining time-based comparative queries with similarity search from generative AI, apps can more easily provide conversational access to personal information.
Read Behind the Development of WHOOP Coach by WHOOP Chief Technology Officer Jaime Waydo to learn how it works.
What Is a Data Time Machine?
Traders on Wall Street have employed time series databases to analyze trading data for years. Now, they’re being adapted for generative AI to answer questions about time, from trading strategy to health trends to donut intake!
A time series database that acts like a Time Machine has five elements:
All data is online, with no need for offline archives,
All data is instantly discoverable and queryable,
Data can be replayed and rewound as required,
All data has a real-time view of the current location and state
Query performance does not degrade as data volume grows
For more, read James Corcoran on these five elements of Vector Databases that act like a Time Machine in Why Your Data Might Need a Time Machine.
Follow the {The, Weekly, Vector} Community!
Check out these Substacks we love:
Follow our growing list of writers,
Have an idea for {The, Weekly, Vector}? DM our editor, Mark Palmer.