Developer Uses DeepSeek to Reverse-Engineer Xiaomi Fitness Calorie Algorithm
10/21/2026 — 10/21, 21:17·1 sources (all)·1 reports (all)
Key takeaways
- A developer used DeepSeek to reverse-engineer Xiaomi's fitness pedometer algorithm, uncovering logic bugs, extreme thresholds, and calculati…
Background & analysis
On October 21, 2026, according to a V2EX report, a programmer who grew suspicious of fluctuating calorie data despite maintaining a consistent daily workout routine used DeepSeek to reverse-engineer the step-counting algorithm of the Xiaomi Sports app when used without a wristband. The analysis revealed that the core logic relies on categorizing steps per minute into distinct tiers: 0.04 calories per step for 138 steps per minute or fewer, 0.06 calories per step for 139 to 277 steps, and 0.10 calories per step for 278 steps or more. The reverse engineering showed that the highest tier requires a frequency equivalent to top sprinters sustained for a full minute, which is practically unattainable for humans, meaning only the first two tiers are effectively utilized. Additionally, the algorithm introduces calculation errors due to rounding during distance computations at each synchronization interval, and suffers from a bug where calories are repeatedly accumulated within the same minute.
AI-generated from 1 reports · updated 16 hours ago
Latest turnA developer used DeepSeek to reverse-engineer Xiaomi's fitness app calorie calculation algorithm when carrying a phone without a band. The analysis revealed that the algorithm relies on simplistic minute-based step brackets, suffers from rounding errors in distance calculation, and contains bugs that repeatedly accumulate calories.
Related AI tools
- Heat index
- 60
- All sources
- 1
- All reports
- 1
- First seen
- 18 hours ago
Reports on this story headlines open the original
A developer used DeepSeek to reverse-engineer Xiaomi's fitness app calorie calculation algorithm when carrying a phone without a band. The analysis revealed that the algorithm relies on simplistic minute-based step brackets, suffers from rounding errors in distance calculation, and contains bugs that repeatedly accumulate calories.
V2EXAI score 69
Other stories people are talking about
- Heat index 743DeepSeek open-sources DeepInfer large-scale inference framework17 sources
- Heat index 96NewJensen Huang on GPU lifespan and DeepSeek inference margins1 sources
- Heat index 511DHS orders federal civilian agencies to purge DeepSeek tech12 sources
- Heat index 650NewDeepSeek closes roughly RMB 100B round with CATL and Tencent backing7 sources
- Heat index 838RisingNvidia confirms Vera Rubin volume production, locks in packaging capacity13 sources
How is heat calculated?About the methodHide
Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.
This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source. If you believe a headline or summary infringes your rights, email the contact address in the footer with the page URL and basis of your claim, and we will remove or replace it after review.
- Surge
- Discussion rising fast
- New
- First report within 6 hours
- Rising
- Still gathering discussion
