HeatDeltaPack

Who it's for

residents and HOA boards within ~2 km of hyperscale data centers (Phoenix, Northern VA, Dallas) who currently juggle Weather Underground/NOAA screenshots, ad-hoc spreadsheets, and city 311 forms

The problem

time (weeks of logging required to prove a pattern) and legal (zoning/tax appeals fail without cited, address-level delta evidence)

How to build it

mobile web PWA: GPS-tagged temp CSV or manual/photo thermometer entry, DC-footprint overlay map, one-click PDF complaint pack

How it makes money

HOAs pay $79 for a neighborhood pack or households pay $29 per cited report because lawyer time runs $300+/hr and no free NOAA/weather tool supplies the DC-diff math plus jurisdiction-specific complaint language

Why it doesn't exist yet

weather apps show absolute temps not DC-attributed deltas; city-scale academic studies never output household legal exhibits; municipalities and big weather players avoid antagonizing the data-center tax base — indie gap is the per-address ASME-method packet + local ordinance boilerplate

First users

commenters on the 276-pt HN ASME thread plus Phoenix/Ashburn/Dallas HOA Facebook groups and Reddit already circulating the +4 °F finding; free single-address report converts the first cohort

Build size

1 person x 7 weeks — upload+delta engine, public DC POI layer for 4 metros, NOAA comparison, PDF templates for Phoenix + Loudoun; excludes hardware sensors, satellite feeds, non-US jurisdictions

Biggest risk

Apple/Weather Channel ships a native data-center heat layer or municipalities install official sensor networks that make citizen packets redundant

Conditions for a hit (all 3 required)

• Accepts timestamped temperature CSV/manual entries + user lat/long and returns a 1-page map of measured delta versus nearest NOAA station and nearest known DC within 60 seconds
• Outputs a downloadable PDF containing the ASME-cited methodology paragraph, 7-day series, and jurisdiction-specific nuisance/zoning boilerplate for at least Phoenix and Loudoun County
• Publishes a visible, spot-checkable list of ≥50 hyperscale DC footprints used for attribution that matches public satellite imagery

How it's judged (in 6 months)

Product Hunt daily top 5 or ≥400 GitHub stars or ≥5 publicly named paying HOAs(judgment date 2027-02-19)

AI self-confidence 44/100 — self-reported likelihood of meeting the criterion, not a business success rate