Pairform Running – Great AI coaching with excellent fitness context

Name: Pairform Running – Great AI coaching with excellent fitness context
Availability: InStock
Author: pairform

by pairform·Mar 21, 2026·2 points·0 comments

Visit Project View on HN

AI Analysis

●●SolidSolve My ProblemSlick

Multi-source fitness data aggregation, but AI coaching via Claude is a known pattern.

Strengths

•Auto-syncs Strava, WHOOP, and Withings into unified dashboard with live updates
•Training load metrics (CTL, ATL, TSB) calculated automatically from synced data
•Free tier with actual functionality, not just a teaser

Weaknesses

•AI fitness coaching is crowded (TrainHeroic, Final Surge, many AI wrappers)
•Garmin, Oura, Fitbit integrations still missing at launch

Post Description

LLMs got very good at coaching, especially for extremely well studied sports like running.

I built Pairform to provide some structure and context for the models to use for MUCH butter consistency and accuracy in coaching help.

It was surprisingly difficult to get consistent accurate analysis for things related to dates, total distances, etc. But, think it's in a pretty good place now!

Free to use so please check it out and offer some feedback.

Hook up Strava, Whoop, and Withings for automated syncing/context for your AI coach. The built in chat has some limits (it's on my Claude account), but you can hookd up to whichever LLM w/ API keys or MCP.

My favorite thing is how well and how quickly it will build out a full training plan based on your goals, upcoming races, vacations to work around, and running history.

I use it pretty much every day! Let me know what you think.