Proprietary & Confidential · Watter™ — do not distribute
01 / 07
Series A · $25M · 2026
WATTER

Heat-aware AI inference.
Cheapest per-token by physics.

GPU servers in hotels and homes. Waste heat captured to deliver free hot water. Compute sold through OASIS — our managed inference platform. Heat-aware routing steers workloads to devices that need thermal output.

Hyatt · Live OASIS · Production USAF · SBIR Phase II 10 Homes · Octopus Energy
Proprietary & Confidential WATTER™ · Series A 2026
MegaWATT · Live Net Zero Cloud
Operating Today Hyatt Reunion · 10 homes live · OASIS shipping
02 · The Stack

One stack. Three layers.

Hardware deployed inside existing buildings. Compute sold as managed inference. Heat-aware routing makes the per-token cost structurally cheaper than any other GPU cloud.

Layer 01 · Hardware WATTER GPU rack hardware
Compute · Captured Heat

WATTER Hardware

GPU racks deployed in hotels, multifamily, district plants. 80–90% waste heat captured into existing DHW loops. Closed dielectric loop · 64× B200 per rack · 100 kW. Behind the meter. Zero new grid load.

  • Hyatt Reunion · production cluster live
  • 10 Apollo units live in homes today
  • PUE 1.05 · ERE 0.210 · validated
Layer 02 · Platform
WATTER | Oasis
+ New Chat
compare MI350 AMD card to other...
How many lives does a cat have
What year is it
Who is your leader
What is avg first hour rating
how many lives does a cat have
Who is your leader
System Prompt
Thinking
I do not have a leader. I am an artificial intelligence developed by Google, and I operate according to my programming, safety protocols, and the instructions you provide.
Who is the CEO of WATTER
System Prompt
The CEO and Co-Founder of WATTER is James A. Hancock. He is a former NASA data scientist and an expert in edge computing.
Managed Inference · Live

OASIS

Managed LLM inference on commodity NVIDIA. Think Groq — but on B200s, with free cooling. OpenAI-compatible API. Open-weight (Qwen, DeepSeek, Llama, Mistral, Kimi). Auto-routing across the fleet.

  • oasis.watter.com · shipping
  • 1-URL drop-in from OpenAI
  • Per-key budgets · audit logs
Layer 03 · Routing Heat-aware routing — residential WATTER unit
Structural Moat

Heat-Aware Routing

Workloads steered to devices that need thermal output right now. Hot-water demand becomes the compute scheduler. We monetize energy that hyperscalers waste.

  • ~40% lower per-rack OPEX
  • Demand-shaped dispatch
  • Cheapest $/token by physics
Proprietary & Confidential WATTER™ · From GPU to BTU · Series A · $25M
03 · Live Today

Operating today.

Live commercial deployment · production inference platform · government contract · residential utility partnership. Post-prototype. Entering deployment scale.

Hyatt Reunion Tower
Live · Operating

Hyatt Reunion Tower · Dallas

NVIDIA B200 production cluster live in the central plant. Closed dielectric loop into DHW. Inference running in production. Per-rack capacity 64× B200 · 100 kW.
B200cluster live
100 kWrack capacity
99.97%30-day uptime
85%heat → DHW
WATTER | Oasis
+ New Chat
compare MI350 AMD card to other...
How many lives does a cat have
What year is it
Who is your leader
What is avg first hour rating
how many lives does a cat have
Who is your leader
System Prompt
Thinking
I do not have a leader. I am an artificial intelligence developed by Google, and I operate according to my programming, safety protocols, and the instructions you provide.
Who is the CEO of WATTER
System Prompt
The CEO and Co-Founder of WATTER is James A. Hancock. He is a former NASA data scientist and an expert in edge computing.
Production · Shipping

OASIS Platform

Managed LLM inference at oasis.watter.com. OpenAI-compatible. Open-weight inference (Qwen, DeepSeek, Llama, Mistral, Kimi).
Scott Air Force Base
SBIR · Phase II

US Air Force · Scott AFB

$1.25M Phase II non-dilutive. Congressional add for $5M/yr recurring defense revenue gaining traction.
MegaWATT pilot mechanical room
Pilot · Validated

MegaWATT Pilot

Live since April 2026 · ~1,480 gal/day DHW. Architecture pre-validated for the 64-rack scale-up at Reunion.
Octopus Energy vans
Partnership · Signed

OctoWATT · Octopus Energy

10 Apollo units operating in homes. OctoWATT is the consumer plan · compensation runs through the homeowner's energy bill via Octopus · residential scale via the utility channel.
Proprietary & Confidential WATTER™ · From GPU to BTU · Series A · $25M
04 · Why We Win

Already validated. Both segments.

11 sites operational today — 1 commercial (Hyatt Reunion) + 10 residential Apollo units. Same architecture, same heat-aware routing, both segments served. Cheapest per-token by physics.

The Moat · Heat-Aware Routing

Workloads route to devices that need heat right now.

Inference jobs are dispatched not just by latency and price — but by thermal demand at the destination. Hotels heating water in the morning rush get the work. Homes pre-heating tanks at peak get the work. The rack pays for itself twice — compute revenue plus displaced gas.

Demand
DHW load profile per site
Route
Inference dispatched to thermal-needy rack
Result
Compute + heat both monetized
Net: ~40% lower OPEX per rack vs hyperscaler colo. Sustainable cost advantage that compounds with every deployed node.
Live in Production · Both Segments

11 sites operational. Already shipping.

1
Commercial · Hyatt Reunion
10
Residential · Apollo

One architecture. Both segments. Span/XFRA just launched the commercial-only version ($419M raised · $103M revenue) — the market validated the model. WATTER is already operating it, with heat capture, across both consumer and commercial. Validated, shipping, and broader.

Deploy Time
Days, not years
Grid Load
Zero net new
Permitting
Behind the meter
IP
US 11,974,412 B2
Proprietary & Confidential WATTER™ · From GPU to BTU · Series A · $25M
05 · Half the CapEx · None of the Wait

$60B traditional. ~$36B distributed.

1 GW of AI compute the old way: ~$60B and a 5–7 year utility queue. WATTER delivers it as 10,000 × 100 kW hotel sites riding mechanical rooms, electrical service, fiber, and hot-water demand that already exist. OPEX savings not even counted below.

Traditional · 1 GW gas-powered DC
GPU$30B
Other IT$11B
Labor$4.7B
Materials$3.9B
Gas Plant$3B
Fitout$2.6B
Elec$1.8B
Mech$1.7B
$60B
5–7 yr wait · build from scratch
−$24B
−40 %
CapEx saved
WATTER · 10,000 × 100 kW hotel sites
GPU$30B
Other IT$4B
Install$2B
~$36B
8 weeks per site · 10,000 in parallel
Time to first megawatt
8 wk vs 5–7 yr
No utility queue. Behind the meter. Sites build in parallel, not series.
What we don't rebuild
$0 infra reuse
Mechanical room · electrical service · fiber · DHW loop. Already in the building.
OPEX bonus (not counted above)
~$30K per site / yr
Closed-loop cooling = $0 water. Waste heat displaces hotel gas. No marginal HVAC.
Sources: Orennia · Stanford · Cushman & Wakefield · WATTER baseline: $3.6M / 100 kW site (see BASELINE_ASSUMPTIONS.md)
Proprietary & Confidential WATTER™ · From GPU to BTU · Series A · $25M
06 · Economics

$1.68M per hotel. 10 sites = $16.8M ARR.

LG hotel deployment: 64× B200 at $3.00/GPU-hr × 8,760 hours. Capacity sold through OASIS before energization.

10-Deployment Forward ARR
$16.8M
Series A funds the rollout. Each site: $3.6M CapEx · ~5 mo equity payback · 20.3% IRR. Pre-sold capacity. Investment-grade counterparties.
10 × $1.68M = $16.8M ARR
SBIR Phase II
$1.25M non-dilutive
Apollo Residential
10 units live · scaling
Hotel Take-Home
$73K / yr / site
Water Saved
876K gal / yr / site
What's Next · Containerized Deployment

From 10 sites to fleet — factory-built drop-in.

MegaWATT Container ships pre-integrated. Same compute, same heat-aware routing — derisked, easier, cheaper at scale. Site-to-revenue compresses from 14 weeks (retrofit) to 14 days (drop-in). Logistics, not construction.

14 daysSite-to-revenue
Factory QCPre-integrated
Drop-inConcrete pad + tie-ins
Proprietary & Confidential WATTER™ · From GPU to BTU · Series A · $25M
07 · Traction

Anchor live. Channels signed.

Commercial · government · residential · industrial. Four go-to-market channels operating in parallel.

Validation · Public
Span / XFRA + NVIDIA
Same distributed compute model, no heat capture. $419M raised. $103M revenue. Validates the category. WATTER is structurally cheaper.
Platform · Production
OASIS
oasis.watter.com live. OpenAI-compatible API. Open-weight inference (Qwen, DeepSeek, Llama, Mistral, Kimi). 1-URL migration.
Proprietary & Confidential WATTER™ · From GPU to BTU · Series A · $25M
08 · The Round · From GPU to BTU

Series A:
$25M.

Funds 10 LG hotel deployments ($16.8M forward ARR), the Apollo residential rollout, and the defense pipeline. Hyatt Reunion is live. OASIS ships in production. Every dollar accelerates already-operating revenue.

James A. Hancock · Founder & CEO
Former NASA Data Scientist · Founder, Parasanti · BloombergNEF 2026 Pioneers Finalist (Data Center Efficiency).
james@watter.com · watter.com · content.watter.works
Proprietary & Confidential WATTER™ · Series A · $25M
Use of Funds
Scale 10 LG hotel deployments $3.6M CapEx per site · 64× B200 each · OASIS-contracted compute · capacity pre-sold.
Ship OASIS at production scale Heat-aware routing across the fleet. Per-key budgets, audit logs, OpenAI-compat API at scale.
Defense Lock USAF recurring Convert SBIR Phase II + congressional add into $5M/yr recurring revenue line. Expand DoD OASIS surface.
Residential Apollo rollout OctoWATT consumer plan with Octopus Energy utility-bill compensation. Distributed inference substrate via 130M US water heaters.
Derisk Containerize deployment Factory-built MegaWATT containers. Same architecture, easier & cheaper at scale. Site-to-revenue in 14 days.
An Investment That Holds Water