Firm power is repricing. Geothermal is the only source that is firm and everywhere.

• Data-center electricity: +17% in 2025, ~945 TWh/yr by 2030 (IEA base case) — and it buys firmness
• Public target: $45/MWh by 2035 (DOE Enhanced Geothermal Shot). Our mission: $30
• Superhot rock (>375 °C): 2.2–2.9× enthalpy per kg vs 250 °C brine; 1–2 producers ≈ a 45–50 MW block

The industry drills hotter than it can see.

• Superhot wells die at transients: the 20-well base rate is thermal stress at quench and kick-off — nobody has measured those hours
• The working tool stack tops out at 200 °C, 150–200 h class. "350 °C" tools are survival ratings for 4-hour runs
• Above 375 °C: no qualification service, no test standard, no ruler. Every envelope negotiated ad hoc

Keep the electronics cold. Measure the hot. Correct in place.

• Passive Inconel waveguide through the hot span; all active components behind a barrier below 100 °C — the IDDP-2 architecture made permanent
• 5-point temperature profile from echo time-of-flight
• In-situ drift correction (reference notch) + fouling correction (dual-depth amplitude ratio) — the two things that kill instruments in practice
• Simulated 500 h @ 450 °C: corrected 2.23 °C vs 9.63 °C uncorrected, gate ±5 °C simulated

The sensor is the product. The yardstick is the company.

Built, run, audited — and labeled.

• Readout engine + digital furnace: 47 pre-registered contracts green; full CAD and manufacturing drawings; bench control stack
• 5/5 inventions virtually qualified against their real bench gate checkers — 56 contracts green simulated
• Claim-traced teardown of the whole tool stack — 34 verified claims, corrections published
• Audited cost model with published QA record — arithmetic verified from the stored trajectories
• Labeled honestly: zero physical runs yet. That is what the round buys

$61.65 → $29.80/MWh. One named invention per rung.

Five inventions. Each with a bench gate and stated kill conditions.

• E0 Cool-Brain — the instrument + the qualification layer · keystone: 500 h @ 450 °C, ±5 °C
• E1 Shift-Latch — commandable reservoir reconfiguration · window 54–137.6 µm simulated
• E2 Clear-Chemistry — keep the hot branch conductive · 3 chemistry classes pass
• E3 Conversion — installed block ≤$1,000/gross-kW · $367/kW gap, named
• E4 Ember-BHCT access + E5 Flex-Seal survival — 300–374 °C ladder, 25-yr duty life

Three numbers. Each converts rooms into capital.

• R1 · ≤30 days — chain proof: the waveguide hears the notches (room temp, <$2k)
• R2 · ≤90 days — first hot data with deliberate drift/fouling attacks + commanded latch release + power-block cost-gap memo
• R3 · Q1 — the 500 h @ 450 °C qualification record (the segment's first, witnessed) + Shift repeatability on 3 well pairs
• SGW week, Feb 8–10: the record's coming-out venue

Three lines of revenue. One ladder to gigascale.

One founder today. Four founding seats. Named artifacts for each.

• Shipped a mining data platform as co-founder/CTO — then built this field's first adversarially-verified evidence base
• Corrected the field's folklore in public (the "45–50 MW target" that doesn't exist) — the results brand
• Open seats: ultrasonics/acoustics · HT packaging/mechanical · qualification/test · drilling-ops advisor
• Founding equity + their name on the paper and on the qualification record