My Journey

The Mission Statement (The "Why")

My entire career has been driven by a single question: how can we build systems that are more efficient, more autonomous, and more equitable? I believe the future of the internet isn't in a few centralized clouds, but in a decentralized network owned by the people who use it. My life's work is to build the physical hardware and software that makes this possible.

"My entire career has been driven by a single question: how can we build systems that are more efficient, more autonomous, and more equitable?"

The Preparation (The "Proof of Work")

Service & Discipline (1995-1999)

My journey began with a foundation of discipline and tactical execution as a Cavalry Scout in the US Army. My time in Bosnia with NATO forces taught me how to operate complex communication systems in high-stakes environments.

Building the Foundation (1999-2008)

After my service, I dove into the building blocks of infrastructure. I wasn't just on a computer; I was mapping the physical world as a GIS specialist, managing county-wide systems, and designing large-scale RF and cellular communication projects.

Enterprise & Scale (2008-2021)

I took this field experience to the enterprise level. At XTO Energy (an ExxonMobil company), I spent six years designing and commissioning control systems across multiple regions. This is where I mastered industrial automation. This deep-stack experience led me to DISH Network, where I was a Leased Executive for Wireless Innovation. I was on the front lines of 5G, researching LEO/MEO/GEO satellite tech, Open RAN ecosystems, and edge computing. I saw the future of wireless, but I also saw its centralized limitations.

The Mission (The "Now")

My time at DISH crystallized my vision. The world didn't need another incremental improvement; it needed a paradigm shift. In 2021, I left to focus 100% on my main focus in life: RevoFi. RevoFi (and its predecessor, RevoKind) is the culmination of my entire journey—blending AI, blockchain, wireless, and edge computing into a single platform. We are building the decentralized physical infrastructure (DePIN) that I saw the world needed. This is more than a company; it's a new model for the internet.

RevoFi: The Network You Own

We are building a decentralized physical infrastructure network (DePIN) that combines Wi-Fi, AI, and cloud computing. Our mission is to democratize connectivity and compute.

Explainer 1: What is RevoFi? (The Analogy)

Imagine if your home Wi-Fi router was also a mini cloud server. Now, imagine you got paid every time someone used it. That's RevoFi. We've built hardware (like the RevoFi Link and Micro) that provides powerful Wi-Fi 6 connectivity and, thanks to its onboard NVIDIA GPU, can also run AI and cloud workloads. Instead of all compute happening in a massive Amazon data center, RevoFi distributes it to the 'edge'—to your homes and businesses.

Explainer 2: The "WaaS" Pivot: A More Powerful Network

Our vision has evolved. We are upgrading the RevoFi network into a 'Workload-as-a-Service' (WaaS) platform. This means our network won't just provide wireless access; it will become a global marketplace for compute. Businesses and developers will be able to buy 'Revos Data Credits' (RDC) to run their applications—like AI models, decentralized CDN, or VPNs—on our network. This is a massive leap forward, creating far more utility and revenue for our device owners.

Explainer 3: The "Dual Reward" System: How You Get Paid

Our new economic model is designed for sustainability. It's simple:

USDC (Real Revenue): When a business pays (in RDC) to run a job on your device, you get paid a share of that revenue in a stablecoin like USDC. This is 'Proof-of-Work'.

RVS (Network Incentive): You also earn our native token, RVS (an ERC-20 token on Base), simply for keeping your device online and available. This is 'Proof-of-Availability'.

Thought Leadership

Building in public. Deep dives into the strategy, economics, and tech behind the WaaS pivot and DePIN's future.

Patents & Intellectual Property

My work is grounded in verifiable, granted intellectual property. This section lists key patents that form the foundation for my work in decentralized systems and network architecture.

The Future of DePIN is Not a Single Function. It's a Marketplace.

Tags: DePIN, WaaS, Strategy, RevoFi, Edge Computing

The decentralized physical infrastructure (DePIN) space is one of the most exciting movements in technology. However, many first-generation DePIN projects are limited by a critical flaw: they are single-purpose networks. They do one thing—be it file storage, wireless, or VPN—and their economic models are tied exclusively to that single function.

This is not a sustainable long-term vision.

That is why I am leading the strategic pivot for RevoFi. We are transitioning from a single-purpose network to a multifaceted, high-margin "Workload-as-a-Service" (WaaS) platform.

This isn't just a minor change; it's a fundamental re-architecting of our business. This pivot is designed to leverage all our existing assets—specifically our central high-performance Gigabyte G292-280 server with its 8x NVIDIA A100 GPUs, and our entire distributed fleet of 158+ deployed edge devices.

What is a Workload-as-a-Service (WaaS) Platform?

In simple terms, we are turning our network into a dynamic, multi-tiered compute marketplace.

Instead of just one service, our hybrid infrastructure will capture revenue from multiple, concurrent compute markets. The platform will be intelligent enough to provision different jobs to the hardware best suited for the task:

• The Central Compute Hub (Our A100 Server): This is our high-performance anchor. It will handle high-intensity, demanding workloads like AI model training and inference, running on NVIDIA's Triton Inference Server.
• The Distributed Edge Fleet (Our 158+ Devices): This is our "far-edge" tier. Its strength is geographic distribution, making it perfect for low-latency tasks like running a Decentralized CDN, providing "light" AI inference for local analytics, or offering network services like dVPN.

Crucially, this model also allows us to de-risk our growth by establishing an immediate revenue floor. We will simultaneously deploy workloads for established DePIN networks like Akash and Bittensor, selling our compute capacity to their markets while we build our own.

This pivot aligns our hardware, our software, and our business model into a single, cohesive, and fundable strategy. The future of DePIN isn't about building one road; it's about building an intelligent, multi-lane highway for any and all decentralized workloads.

Incentives vs. Revenue: Fixing DePIN Economics with a "Dual Reward" Model

Tags: Tokenomics, DePIN, RVS, USDC, WaaS, RevoFi

Many DePIN projects have a fatal economic flaw: they confuse incentives with revenue. Paying people in an inflationary token to simply turn on a device is a customer acquisition cost, not a business model. A sustainable economy requires real, paying customers.

When I redesigned the RevoFi economy, I made it my top priority to fix this. That's why our new WaaS platform is built on a "Dual Reward" System. This model creates a clear and transparent separation between the two forces that make a network grow and thrive.

Here is how it works.

1. RVS (The Incentive): This is "Proof-of-Availability"

• What it is: The Revos (RVS) token is our incentive mechanism. It's an ERC-20 token on the Base (L2) network.
• How you get it: All online and healthy device owners receive RVS from our on-chain vesting contract.
• Why you get it: You get RVS simply for being connected to the WaaS platform and being "ready" to accept work. This is our "Proof-of-Availability". It’s the reward for helping us build and secure a globally distributed network, fulfilling our original 50-year halving promise and incentivizing network scale.

2. USDC (The Revenue): This is "Proof-of-Work"

• What it is: This is your share of real, paying customer revenue.
• How you get it: When a developer or business pays our platform to run an actual, paid job—like an AI inference task or a CDN request—the specific device owner who completed that work gets paid.
• Why you get it: You get paid in a stablecoin (like USDC) because you performed a billable service. If a customer spends $100 on a job, a portion of that $100 is paid directly to the device owner who did the work. This is "Proof-of-Work" in its purest form.

This dual system creates a sustainable, real-world business model. The RVS token incentivizes the network's potential, while USDC pays for its actual performance. This is how we build an economy that lasts.

Purging the Legacy: Why I'm Moving RevoFi's Settlement Layer to Base (L2)

Tags: Strategy, Blockchain, Base, L2, RevoFi, ERC-20

As an architect, you must be willing to tear down a flawed foundation to build something stronger. As part of our pivot to a WaaS platform, I have made a definitive strategic decision: we are purging the legacy, forked NKN blockchain (rvs-1.0.2) and all its associated technical debt.

Maintaining a proprietary L0/L1 blockchain is a massive, expensive, and unnecessary distraction. It is a drain on resources that should be focused on our core product: the WaaS platform itself.

That is why we are migrating our entire settlement layer to Base (L2).

This move is a hard-line in the sand, and it provides three immediate, non-negotiable advantages:

1. Elimination of Maintenance: This move "eliminates all blockchain maintenance overhead". We no longer have to worry about consensus, security, or core protocol-level bugs. We can focus 100% on our application.
2. Instant Security & Interoperability: We immediately inherit the world-class security of Ethereum and the scalability of an L2.
3. The Coinbase Ecosystem: Building on Base "unlocks the Base/Coinbase ecosystem" for grants, investment (e.g., Y Combinator), and, most importantly, seamless user onboarding.

The Revos (RVS) token will be re-launched as a standard ERC-20 token on Base. Our 50-year halving tokenomics will be immutably programmed into a new, auditable, on-chain smart contract called the RVS_Vesting_Vault. This is far more transparent and trustworthy for our community than a complex, custom-built consensus protocol.

We are done managing a legacy blockchain. We are now a modern, capital-efficient, and high-growth DePIN application building on the best foundation available.

RDC is Not a Token. It's the Key to Enterprise Adoption.

Tags: WaaS, RDC, Billing, Strategy, DePIN, RevoFi

One of the biggest barriers to enterprise adoption of DePIN is economic. A B2B customer cannot and will not pay for services with a volatile token. They need predictable, stable, and simple pricing—just like they get from AWS, GCP, and Azure.

To solve this, we created the Revos Data Credit (RDC).

I want to be perfectly clear: RDC is not a blockchain token.

RDC is our internal, non-speculative unit of account for all services on the RevoFi platform. It is the abstraction layer that separates our customers from the complexity of crypto.

Technically, RDC is simply a decimal-value column in our WaaS platform's PostgreSQL database. It is an internal credit, nothing more.

Here is how it works:

• Customers On-Ramp: A developer or business comes to our portal and buys RDC. They can use a credit card (via Stripe), Bitcoin (via our Lightning node), or stablecoins (like USDC).
• Customers Spend: They then spend their RDC balance to pay for any of our 16+ workloads—compute, storage, connectivity, AI inference, etc..
• Predictable Pricing: We price all our services in RDC, with the goal of being approximately 50% cheaper than traditional cloud providers.

This model gives our customers exactly what they need: a simple, stable, and familiar billing experience. They get the cost savings of DePIN without the volatility and complexity of crypto. This abstraction is the key that unlocks the B2B and enterprise market.

The Economic Flywheel: How RDC Creates Real Demand for RVS

Tags: Tokenomics, RVS, RDC, DePIN, WaaS, RevoFi

In my last two posts, I explained our "Dual Reward" system:

• RVS is the incentive token ("Proof-of-Availability").
• RDC is the billing credit (what customers buy).

Now, let's tie it all together. This is the most important part of our new economy: how we create intrinsic, utility-based demand for the RVS token.

We achieve this by building a powerful, self-sustaining economic flywheel.

The mechanism is a simple, powerful discount. We call it the "Token Sink".

Here is the scenario:

A developer needs to buy 1,000,000 RDC to run their application. They go to our payment portal and see their options:

• Pay with Fiat (Stripe): $100
• Pay with USDC (Base): $100
• Pay with RVS (The Discount): $90

By offering a 10% discount to customers who purchase RDC using our RVS token, we create an immediate and powerful economic incentive.

Why would a customer not take the 10% discount?

This simple mechanic creates a perpetual "token sink." It drives platform users to go to the open market, acquire RVS tokens, and then use them to pay for services. This process takes RVS out of circulating supply and locks it in our treasury.

This is the holy grail of tokenomics. The demand for RVS is no longer based on speculation. It is driven by the real-world utility of our WaaS platform. The more customers we have, the more RDC they buy. The more RDC they buy, the more demand there is for RVS to get the discount.

This is how our two-asset model—separating incentive from utility—fuels its own growth.

Our Server Strategy: How We Monetize a Single NVIDIA A100 Server in 3 Ways at Once

Tags: NVIDIA, A100, MIG, Akash, Bittensor, AI, DePIN

Our platform is not just an idea; it's anchored by serious, high-performance hardware. The core of our "Central Compute Hub" is a Gigabyte G292-280 server packed with 8x NVIDIA A100 80GB GPUs.

The challenge isn't having the hardware; it's monetizing it effectively. How do you run multiple, demanding, and isolated jobs on the same server at the same time?

The answer is NVIDIA Multi-Instance GPU (MIG) technology.

MIG is a game-changer. It allows us to partition each physical A100 GPU into multiple, smaller, hardware-isolated GPU instances. This is critical because it provides a guaranteed Quality-of-Service (QoS) and predictable latency for each "tenant" running on the server.

With MIG, we can execute our multi-workload strategy and create a diversified revenue stream from day one. Here is our initial allocation:

1. Proprietary AI Platform (Our Future): We allocate the largest portion of our GPU resources (e.g., 6 MIG instances) to our own high-margin "Edge AI Development" workload. This is where we run the NVIDIA Triton Inference Server to power our proprietary AI services.
2. Akash Provider (The Compute Marketplace): We simultaneously allocate 2 MIG instances to act as a provider on the Akash network, selling our compute to their open marketplace.
3. Bittensor Miner (The AI Network): We allocate another 3 MIG instances to run miners for Bittensor (specifically Subnets 19 and 21).

This multi-track model is a pragmatic and powerful revenue hedge. The immediate, stable cash flow generated from providing services to Akash and Bittensor helps cover our operational costs (power, co-location).

This allows us to fund the development and scaling of our own, higher-margin AI platform. We don't have to wait. We are monetizing every part of our hardware right now.

The 'AI Compute Continuum': How Our Edge Devices and A100s Work Together

Tags: AI, Edge Computing, NVIDIA, Jetson, A100, gRPC

Our infrastructure is a hybrid. It has two distinct tiers: the "Central Compute Hub" (our powerful NVIDIA A100s) and the "Distributed Edge Fleet" (the 158+ devices running NVIDIA Jetson modules in homes and businesses).

A common question I get is: how do these two tiers work together?

The answer is what I call the "AI Compute Continuum." We are implementing a "Model B: Full Inference Offload" architecture, which intelligently balances the workload by offloading the heavy computation from the edge to the core.

This model gives us the best of both worlds: the low-latency presence of the edge device and the raw power of the central server.

Here’s a practical example of how we power a "Smart Retail" analytics service:

• The Edge Device (Jetson Nano): Our device, sitting in a retail store, uses its camera to run a lightweight DeepStream (video) client. Its only job is to perform hardware-accelerated capture, decode the video (using its onboard NVDEC), and pre-process the frames. It does not run the full, complex AI model locally. This keeps its resources free.
• The Transport (gRPC): The device batches these pre-processed video tensors and sends them over a secure gRPC connection to our central server.
• The Central Server (A100s/Triton): Our server, running the NVIDIA Triton Inference Server, hosts the complex, pre-trained AI models (like YOLO for object detection). It receives the gRPC stream, performs the heavy-duty inference, and sends a tiny, lightweight result (e.g., {"person_count": 12, "queue_wait_time": "3 min"}) back to the device or customer portal.

This "continuum" is incredibly efficient. It solves the problem of running complex AI at the "far edge" by not doing it. We let the edge device do what it's good at (capture and pre-process) and let the central server do what it's good at (heavy-duty AI inference). This is the scalable, hybrid architecture that will power our next generation of AI services.

Architecting the 'Phoenix App': The New Command Center for RevoFi

Tags: PhoenixApp, Mobile, Strategy, WaaS, RevoFi

As part of our strategic pivot, I conducted a full analysis of our existing mobile app. My conclusion was firm: "Scrap the entire existing mobile app".

The legacy app was a security risk, technically obsolete, and functionally misaligned with our new WaaS (Workload-as-a-Service) model.

We are building its replacement, the "Phoenix App," completely from scratch.

This new app is no longer just a "wallet." It is the "Command Center for your WaaS DePIN Platform". It is a modern, Flutter-based application built around three core modules that represent our new business model.

• Module 1: The Fleet Manager (The "Devices" Tab): This is the new heart of the app. It's a real-time (WebSocket-powered) dashboard for your hardware. You can see all your mapped devices, their status ("Online," "Idle," "Working"), and their health telemetry. It will feature a new, secure QR-code-based onboarding flow to map a device's machine_id to your user account.
• Module 2: The Billing & Payments Hub (The "RDC" Tab): This is the new commercial engine. This tab displays your "Revos Data Credit" (RDC) balance—which, as I've noted, is a database credit, not a token. This is where you will "Top Up" your RDC balance using our multi-rail payment system: Stripe (credit card), Base Pay (USDC), or our BTC Lightning node.
• Module 3: The Service Control Center (The "Services" Tab): This is the new, immediate value proposition for device owners. This is a UI to control the services running on your hardware. From here, you will be“bleto toggle your dVPN and firewall, use a conversational AI chat interface, or even one-click deploy a containerized, self-hosted private messaging server (like Matrix-Synapse) on your own device.

The only "golden thread" we are keeping from the old app is the existing Firebase Auth project. By reusing it, our 72+ current users can log into this brand-new app with their old credentials, ensuring a seamless migration.

The Phoenix App is the new face of RevoFi. It's a secure, powerful, and modern tool built for the utility-first WaaS platform we are launching.