VENDOR.Max · AI Edge Infrastructure TRL 5–6 2.4–24 kW

Grid-Independent Power
for AI Edge
Infrastructure

Designed for AI edge nodes, distributed compute sites, and infrastructure where power availability constrains deployment.

VENDOR.Max is being developed as an infrastructure power node engineered for environments where grid instability or absence creates a hard constraint on AI edge deployment density.

TRL 5–6
Technology Readiness Level — laboratory validated
1,000+
Cumulative operational hours — documented
532 h
Longest single continuous operational cycle
10 kW
Mid-range design target — typical AI edge node load
Interpretation note: VENDOR.Max is described at the architecture and engineering design level. All operational characteristics represent design targets at TRL 5–6 (pre-commercial validation stage). A startup impulse is required to initiate the electrodynamic regime; autonomous operation thereafter. This page does not constitute a product claim, performance guarantee, or confirmed deployment offer. Patent: WO2024209235 · ES2950176 (granted).
Autonomous power architecture for grid-constrained AI edge infrastructure — VENDOR.Energy
Interpretation note
TRL 5–6
Validation stage Active development.
DNV/TÜV verification
planned Q2–Q3 2026.
Technology status · Design stage context

VENDOR is an open electrodynamic system operating within classical physical laws. The technology is developed under a TRL-based validation framework, with performance claims gated by independent laboratory testing and certification milestones. Environmental interaction is treated as a coupling medium rather than an energy source.

This page describes a technology at TRL 5–6 under active development. VENDOR.Max is not a certified commercial product and should not be interpreted as a product claim, performance guarantee, or confirmed deployment offer.

All capability statements reflect intended design parameters and documented laboratory operating behavior. Independent verification by DNV/TÜV is the next scheduled milestone (Q2–Q3 2026). Engineering and commercial parameters remain subject to change.

Patents ES2950176  · Granted WO2024209235  · PCT EP23921569.2  · 37 countries CN202380015725.5 IN202547010911
Section 01 · The Problem

AI Infrastructure Has
an Energy Problem

AI compute deployments are scaling faster than the power infrastructure designed to support them. The constraint is no longer silicon — it is energy.

Problem 01
3–7yr
Grid interconnection wait — major markets
The Grid Queue
Grid interconnection timelines in major markets have extended significantly, driven by surging demand from data centers, industrial load growth, and infrastructure backlogs. In several US and European markets, new connection approvals are running years behind deployment schedules. Hardware is ready. Power is not.
Grid Backlog Interconnection Queue Capacity Gap
Problem 02
×4+
Lifetime OPEX multiplier — diesel vs. CAPEX
The Diesel Fallback
Diesel generation remains the default alternative — but it introduces fuel logistics, emissions liabilities, and a cost structure that compounds over time. The longer the grid queue, the deeper the diesel dependency — and the harder it becomes to exit.
Fuel Logistics Emissions Liability OPEX Spiral
Problem 03
0net
Carbon commitments vs. diesel reality — direct conflict
The Carbon Conflict
The largest AI operators hold public carbon-neutral commitments. Diesel backup generation is a direct conflict with those obligations. For publicly listed operators, this is not an operational inconvenience — it is a regulatory and reputational exposure that grows with every generator-hour logged.
ESG Risk Carbon Exposure Net-Zero Conflict
Problem 04
1spof
Single point of failure — grid-dependent infrastructure
Grid as Single Point of Failure
The Iberian blackout (April 2025) and the Berlin Lichterfelde arson (January 2026) — which left 45,000 households and 2,000 businesses without power for up to four days — confirmed that centralized grid infrastructure remains fragile under stress. Grid dependency means one failure vector controls everything.
Grid Fragility Resilience Gap Critical Infrastructure
Problem 05
€/kWh
Energy price volatility — P&L forecasting broken
Energy Cost Volatility
AI infrastructure economics depend on predictable energy cost. Post-2022 European market volatility made long-term P&L modelling for grid-dependent operations structurally unreliable. Operators who cannot fix their energy cost cannot fix their unit economics — and cannot commit to stable pricing for inference or training services.
Price Volatility Unit Economics P&L Exposure
Problem 06
opt
Location driven by grid availability — not strategy
Geographic Lock-In
Operators can build only where the grid has available capacity — not where latency, land cost, cooling, or proximity to users is optimal. Grid dependency is a strategic constraint on where AI infrastructure can exist at all. Edge deployments and distributed inference are blocked before they start.
Site Constraint Edge Deployment Strategic Lock-In
The Result
AI edge deployments, distributed inference nodes, and modular compute infrastructure are increasingly constrained — not by hardware availability, but by the inability to power them reliably, predictably, and independently of grid timelines, fuel logistics, and carbon obligations.
Section 02 · The Solution

VENDOR.Max — Autonomous Power Architecture
for AI Edge and Distributed Compute

Designed as a solid-state autonomous power architecture operating without conventional fuel logistics and without battery replacement cycles — intended to address the full stack of energy constraints facing AI edge infrastructure.

Definition

VENDOR.Max is a modular autonomous power system intended to operate as an autonomous, supplemental, or resilience power layer for AI edge clusters, distributed inference nodes, and modular compute infrastructure. The designed output range is 2.4–24 kW per unit, scalable through modular deployment.

TRL Status

The system is currently at TRL 5–6. The operating regime has been documented in internal laboratory testing, including 1,000+ cumulative operating hours and a 532-hour continuous cycle. Independent verification by DNV/TÜV is the next scheduled milestone, planned for Q2–Q3 2026.

Architecture Note

VENDOR.Max operates through controlled discharge-based interactions within a stabilized nonlinear electrodynamic regime. Gas serves as the interaction medium — not as an energy source. The system requires external input to initiate and sustain operation. It is not a perpetual motion device and does not claim to produce energy from nothing. Detailed operating principles are described on the How It Works page.

2.4–24kW
Designed output range per unit
1,000+
Cumulative operating hours documented
532h
Continuous cycle recorded
TRL5–6
Current validation stage

Designed Deployment Modes

Mode 01
Edge / Distributed Compute Power

Designed to function as an autonomous power layer for AI edge nodes, remote inference clusters, and distributed compute sites where grid connection is unavailable, delayed, or unreliable. Addresses the Grid Queue and Geographic Lock-In constraints directly.

Mode 02
Supplemental Capacity Layer

Designed to add autonomous power capacity alongside existing grid or generator infrastructure — reducing grid dependency and fuel logistics without full infrastructure replacement. Enables a phased transition away from diesel dependency.

Mode 03
Resilience Layer for Critical Compute

Designed for mission-critical compute environments where continuity of power is a hard requirement and conventional backup systems introduce unacceptable operational complexity. Eliminates grid as a single point of failure.

How VENDOR.Max Addresses Each Problem

Problem 01
The Grid Queue

VENDOR.Max is designed to operate without grid connection — as a primary autonomous layer or supplemental capacity. Deployment is not gated by interconnection timelines.

Problem 02
The Diesel Fallback

Solid-state architecture with no fuel logistics and no combustion. Designed to eliminate the OPEX spiral of diesel dependency — no refueling, no emissions liabilities, no fuel theft exposure.

Problem 03
The Carbon Conflict

No combustion, no fuel-based emissions at the point of operation. Designed as a non-fossil-fuel power architecture — intended to be compatible with net-zero infrastructure commitments.

Problem 04
Grid as Single Point of Failure

Autonomous operation independent of grid availability. Designed as a resilience layer that removes grid dependency from the critical failure path — compute continues when the grid does not.

Problem 05
Energy Cost Volatility

No fuel input means no exposure to fuel price markets. Designed to provide a fixed-cost autonomous power layer — enabling predictable unit economics independent of grid tariff volatility.

Problem 06
Geographic Lock-In

Grid-independent architecture means deployment is no longer constrained by grid capacity at a given location. AI infrastructure can be placed where strategy requires — not where the grid permits.

Section 03 · Validation

What Has Been Validated

Operating Record · TRL 5–6
  • 1,000+ hours

    Cumulative operating hours documented in internal laboratory testing, including a 532-hour continuous operating cycle.

  • 2.424 kW

    Current designed output range per unit, scalable through modular deployment.

  • Solid-state

    No combustion, no fuel logistics, no battery replacement cycles — as an intended design parameter.

Patent Portfolio

The portfolio includes a granted Spanish patent (ES2950176), a PCT application with completed international phase (WO2024209235), and additional national and regional filings under examination. Patent protection covers the core operating architecture and electrodynamic regime.

ES2950176 Granted
WO2024209235 PCT · International phase complete
EP23921569.2 EPC member states · Under examination
CN202380015725.5 Under examination
IN202547010911 Under examination
US national phase Under examination
Next Milestone
DNV / TüV

Independent verification is planned. This process will cover the operating regime under independent conditions. Results will be published in the technical Data Room.

Design Partners participate in the verification process directly — with access to methodology, data, and results in real time.

Interpretation note: This page does not claim that the operating physics have been independently validated. That is the purpose of the planned DNV / TüV process. All statements reflect documented internal testing at TRL 5–6.
VENDOR.Max laboratory electrodynamic system — internal operating record
Laboratory documentation VENDOR.Max operating system — internal testing environment. 1,000+ cumulative hours documented. DNV / TüV independent verification planned.
Section 04 · Technical Questions

Questions We Hear
From Technical Teams

Question 01
This sounds physically implausible. How should it be interpreted?

VENDOR.Max does not operate on ambient energy or claim to extract power from nothing. The system uses gas as an interaction medium within a controlled electrodynamic regime — not as a fuel or energy source. External input is required to initiate and sustain operation. The operating architecture is documented in a granted patent (ES2950176) and additional regional and national filings under examination. A patent is not a performance claim — it reflects an examined assertion of technical novelty within the patent system.

Not Free Energy External Input Required Patent ES2950176
Question 02
We need evidence, not narrative.

The operating regime has been documented through 1,000+ cumulative hours of internal laboratory testing, including a 532-hour continuous cycle. This is an operational record, not a proof of concept sketch. Independent verification by DNV/TÜV is the next scheduled milestone — Q2–Q3 2026. Design Partners participate in that process directly. Technical data is available in the Data Room for qualified applicants.

1,000+ Hours 532h Continuous DNV/TÜV Q2–Q3 2026
Question 03
We cannot build infrastructure on TRL 5–6.

TRL 5–6 with 1,000+ documented operating hours indicates a system with a recorded operating regime under controlled conditions. The remaining path to deployment includes manufacturing engineering, independent verification, certification, and scale-up. Design Partners who engage now participate in defining the technical specification. Those who engage later receive what was specified without them.

TRL 5–6 Design Partner Access Specification Input
Question 04
We do not have internal expertise to evaluate electrodynamic systems.

That is precisely the function of independent DNV/TÜV verification. You do not need internal expertise in nonlinear electrodynamics. You need a verification report from an organization your board, your investors, and your insurers already trust. That report is what the Q2–Q3 2026 process produces. Design Partners are present when it does.

DNV/TÜV Independent Verification No Internal Expertise Required

More detailed technical questions are addressed in the Data Room — available to qualified Design Partner applicants under NDA.

Access Data Room
Section 05 · Design Partner Program

Design Partner Access —
Current Cohort

VENDOR.Max is in active development toward a rack-compatible form factor for AI edge and distributed compute environments. A closed Design Partner cohort is open for qualified infrastructure operators and technical evaluators.

Maximum cohort size: 3 organizations. This is an engineering constraint, not a marketing device. Design Partner participation requires direct collaboration with the engineering team on technical specification. More than three concurrent partners creates specification conflicts and reduces the quality of engagement for all.
3
Maximum
Design Partner
slots
Cohort Open
What Design Partner Status Includes
  • Technical Specification Input

    Participation in forming the rack module spec — integration parameters, form factor, and electrical characteristics shaped around your infrastructure, not a generic standard.

  • Priority Production Position

    Priority position in the first production series, subject to Design Partner agreement terms.

  • DNV/TÜV Verification Access

    Direct participation in the Q2–Q3 2026 verification process — right to put technical questions to the engineering team and access results in real time.

  • Technical Data Room

    Operational records, patent portfolio, and engineering reports at the scope appropriate to current TRL. Full documentation released after TRL 7–8 under NDA.

  • LOI / MOU on Pilot Deployment

    Terms fixed before series production begins.

What Design Partner Status Does Not Include
  • Public Endorsement or Disclosure

    Engagement is governed by NDA. No public obligations at any stage until you choose to proceed.

  • Commitments Before Evaluation Is Complete

    No binding agreements are signed until your technical team has answers to all questions.

  • Full Technical Documentation at TRL 5–6

    Core engineering documentation is released progressively, gated by TRL milestones and NDA. Full access after TRL 7–8.

Specification Window

DNV/TÜV verification is planned for Q2–Q3 2026. The technical specification window closes before verification begins — an engineering freeze is required to run the process. Organizations that engage as Design Partners before that point shape the specification.

After Cohort Closes

After the current cohort closes, a standard waitlist applies — with no guaranteed position and no influence on specification. Organizations that engage after receive what was specified without them.

3 Design Partner slots. Cohort open now. Specification window closes before Q2–Q3 2026 verification begins.

Section 06 · Economics

Economic Case for
Autonomous Edge Power

The economic case for autonomous solid-state power is driven by what it removes from the lifetime cost of infrastructure — not just what it adds.

What Conventional Power Costs You
  • Fuel procurement and delivery logistics
  • Generator maintenance cycles
  • Battery replacement schedules
  • Emissions compliance overhead
  • Grid tariff volatility exposure
  • Site access visits for refueling
Why the Difference Is Structural

For AI edge deployments — where sites are often remote, access is constrained, and operational continuity is a hard requirement — the lifetime cost difference between conventional backup power and an autonomous architecture is structural, not marginal.

Conventional systems carry low upfront cost but high, compounding OPEX. Autonomous solid-state architecture inverts that model: higher initial investment, near-zero fuel and maintenance OPEX, predictable long-term cost structure.

The further the site from grid infrastructure — and the more critical the uptime requirement — the stronger the autonomous power economics become.

Diesel Generator
Battery + Grid
VENDOR.Max (designed)
Fuel cost
High — continuous
Grid tariff exposure
None designed
Fuel logistics
Regular delivery required
None
None designed
Maintenance
Engine servicing cycles
Battery degradation
Minimal — designed
Replacement cycles
Engine rebuilds
Battery swaps — periodic
Designed for 20+ yr service life
Emissions liability
Direct — combustion
Indirect — grid source
None at point of operation
OPEX predictability
Volatile — fuel price dependent
Partially volatile
High — no fuel input
Site-Specific TCO and LCOE Analysis

A working financial model built against your actual parameters — not a generic benchmark. Available for qualified Design Partner candidates as part of the Strategic Evaluation Package.

Cluster power requirements Current energy cost Geographic constraints Grid situation Payback estimation Lifetime savings envelope
Apply for Design Partner Access

Analysis available to qualified applicants only. Not a public benchmark.

Section 08 · Next Step

Request
Design Partner Access

Design Partner engagement is open now for qualified infrastructure operators and technical evaluators.

What Happens After You Apply
  • 1
    We review your infrastructure context

    We confirm fit based on your deployment scale, geographic constraints, and infrastructure requirements.

  • 2
    Technical session with the engineering team

    No marketing presentation — a working conversation about your infrastructure and the system architecture. Your technical team asks the questions they need answered.

  • 3
    Access to the technical Data Room

    Operational records, patent portfolio, and engineering documentation — at the scope appropriate to current TRL, under NDA.

  • 4
    Design Partner agreement and specification work

    If both sides confirm fit, we formalize the agreement and begin shaping the rack module specification around your infrastructure parameters.

No public obligations at any stage
No commitments before your evaluation is complete

Current Cohort Status

3
Design Partner slots Maximum cohort size — engineering constraint
Specification window closes before DNV / TüV verification begins. Cohort open now.
TRL

VENDOR.Max is at TRL 5–6. This is not a certified commercial product. Design Partner engagement is a technical evaluation process.

NDA

Full technical documentation released after TRL 7–8 under NDA. Data Room access covers operational records and patent portfolio.

IP

MICRO DIGITAL ELECTRONICS CORP SRL, Romania, EU. Patents: ES2950176  · WO2024209235