Felicia Farris

@me8_x1782 I actually really like this suggestion. I’ll definitely discuss this with Jeff. we’re also in the process of adding a new engineer to the team which will help us speed everything up. appreciate the support.

Hi Felicia, Shelby and Volm team, I really like the direction you're taking with Volm - turning idle hardware into productive capacity is a massive problem worth solving. As someone following the project closely, I understand why you’re keeping the core (Edge Compiler, Bare-Metal ZK proofs, and PoEV implementation) closed source for now. That makes complete sense to protect your real technical moat. ⚠️ However, many in the community are still hesitant because there’s currently very little public proof that the system actually works beyond slides and explanations. Would you consider open-sourcing just a small, safe part of the stack? For example: The on-chain ProofBundle Verifier Smart Contracts (Solidity) The ProofBundle data format + a simple Rust verifier library (or at least an example JSON + validation logic) The MetaVault settlement contracts This would allow the community and potential partners to: Verify that proofs can actually be settled on-chain. See that the reward mechanism is real and auditable. Build more confidence without exposing your core IP. Many successful DePIN projects (like Akash and Render in their early days) took exactly this approach - open-sourcing the on-chain and verification layers while keeping the heavy edge/hardware code private. I believe this small step would significantly increase trust and attract more serious supporters, without compromising your competitive advantage. Would love to hear your thoughts. Best regards

#1 The Idle Hardware Problem & Sidecar Solution I spend way too many nights looking at Grafana dashboards. If there’s one thing that actually stresses me out, it’s seeing idle CPU cycles. In software, we have auto scaling. In the physical world? We have a $1.4 trillion waste.

Our Rewards Bridge Layer is the core of this. Instead of fighting with other networks like Helium or Render, we harvest those native rewards and bridge them to Base. We’re not replacing the ecosystem; we’re optimizing it.

This is why we’ve been focused on building @VolmLabs. We’re not just another token; we’re a hardware rooted protocol designed to capture that idle capacity and turn it into real world value

going from bare metal code straight to an automated defi vault in one pipeline is just fascinating. to dig into the zk math or edge compiler, check docs.volm.ai

finally at the top it all surfaces in the automated defi yield strategy apps. the user literally just opens an app and watches their factory robot print money while it sits in the corner. no complex wallet management, no seed phrases, no bridging headaches.

i wanted to actually show the exact stack we built at volm to make this happen. this is our core data-to-yield pipeline and honestly it is pretty wild how these layers stack up. 🧵⤵︎

Let’s talk real constraints. Most "Edge" nodes are just Linux servers disguised as DePIN. Real IoT isn't a Raspberry Pi with 8GB RAM. It’s a Cortex-M4 chip with 192KB. Standard crypto libraries crash these devices instantly. That’s why we built the VOLM Edge Compiler. We stripped the validation logic down to its mathematical bones. If it requires a server rack to validate data, it’s not DePIN. It’s just cloud computing with extra steps. #EmbeddedSystems #EdgeComputing #Engineering

AI models and autonomous systems are currently eating garbage data. That is why they hallucinate or fail. They simply lack ground truth. The missing piece is verified context. #VOLM nodes provide cryptographically signed proofs of physical reality. We validate temperature, precise location, and robotic activity directly from the source. Clean data is the fuel for the autonomous era. We are the refinery. #DePIN #Robotics #AI #base

Morgan Stanley projects Apple could make $133 billion a year on humanoid robots by 2040.

All eyes on robotics

The new humanoids are impressive. It's clear the hardware problem for embodied AI is getting solved, but the real bottleneck remains the software. That classic robotics stack (separate perception, planning, control) is too brittle. The entire field is now shifting to end-to-end (E2E) models: raw pixels in, motor commands out. This is the only path to true zero-shot physical tasking, like "clean up that spill." Hardware is becoming a commodity. The race is on for the general-purpose brain.

VOLM bridges environmental data to @base, turning proof into value. #DePIN #Base #OnchainData #AIoT