I architected and built the core infrastructure that runs AVAP Sphere.
Now, as CTO of 101OBEX, I'm evolving it into an AI-driven platform.
Most infrastructure is built around constraints someone else imposed. I've always been more interested in the constraints nobody has challenged yet — the ones so deeply embedded in how we build systems that people stopped seeing them.
That's why AVAP isn't just a framework. It's a virtual programming language designed for live API orchestration. It runs on PLATON, a kernel that allows us to treat business logic as Virtual APIs: code that can be updated in real-time without redeploys. This logic doesn't just sit in a silo; it breathes across the AVAP Collaborative Network, a physical, decentralized infrastructure where execution power is contributed by the community to ensure true resilience.
The virtualization layer, the physical P2P execution network, the AI orchestration engine: none of those could exist without first solving the problem that everyone else ignores: API logic today is statically deployed, which means it cannot evolve in real time. I believe the next generation of infrastructure will be defined not by who has the best cloud, but by who controls the abstraction layer between intent and execution. That's the space I'm building in.
From language design to physical P2P networks — each project solves a layer of the same challenge: How do you govern autonomous API intelligence that can self-evolve and patch at scale, without the friction of traditional redeploys?
A syntax-agnostic virtual kernel designed to host virtual programming languages. It provides the secure, resource-constrained Bytecode VM and the sandboxed runtime required for high-performance API execution.
AVAP is a virtual programming language purpose-built for API logic: its commands are defined in a central Definition Server, not embedded in the code. Developers write intent; at runtime, the language server resolves each command to the latest definition — enabling Atomic Security Patching across the entire network without redeploys or service interruption. The programmer never worries about what's behind the command.
A live, decentralized P2P infrastructure of AVS nodes and Language Servers. The network is functional and ready — built to provide distributed execution power for the AVAP ecosystem. Now waiting for its first wave of node operators. Target for CNCF proposal.
The orchestration plane where Autonomous AI Agents manage API lifecycles. It defines logical routing, complex flows, and wrappers for legacy protocols (REST, SOAP, gRPC) integrated into the AVAP ecosystem.
A fine-tuned language model with RAG retrieval, purpose-built for the AVAP stack. Acts as the cognitive engine for autonomous agents, enabling routing decisions, self-correcting code, and goal-driven API evolution.
Leading the strategic development of the AVAP Technology Stack. Responsible for systems architecture, open-source governance, and AI-powered infrastructure design across the full product line.
Overseeing initiatives for open standards, education, and the advancement of the AVAP language ecosystem within the non-profit structure.
Participating in organizational working groups and collaborating on cloud-native and API standards as CTO of 101OBEX Corp. AVAP, the technology we created and developed, holds corporate memberships in these foundations, enabling direct access to standards committees and governance forums.
The PLATON Foundation and AVAP Framework are built in the open. These repositories represent the public layer of the ecosystem — the language kernel, execution engine, and definitions server — available for inspection, contribution, and adoption.
Open to conversations about API governance, distributed systems, open-source collaboration, and the future of AI-native infrastructure.