Who is Enfuse.io?
Enfuse.io is a sovereign AI platform company that builds on-prem LLM orchestration and App Factory solutions for regulated enterprises. Founded on the principle that AI platforms should be owned—not rented—we deploy air-gapped, compliant AI systems for defense, government, healthcare, and financial services.
Team
Jacque Istok
Founder & Managing Director, AI Transformation
Jacque leads Enfuse with one mission: make real, usable AI accessible to every organization — not just the Fortune 50. He's spent 20+ years building high-performance data systems and shipping production technology.
Previously shaped data and AI strategy at VMware, co-founded Professional Innovations, and now advises startups across the Northwest. CS + Philosophy at RPI — he cares about how systems work and who they serve.
Cahlen Humphreys
Co-Founder & Chief Product Architect, Sovereign Application Platform
Cahlen builds sovereign edge AI — intelligent agents that run locally, in real time, and never phone home. 10+ years in high-throughput distributed systems, focused on emotionally aware, on-device decisioning at sub-100ms latency. Spoken at NVIDIA GTC, SpringOne, and Datacon LA. Core belief: your AI should be fast, local, and yours.
Jess Barnett
Head of Platform
Jess leads platform and architecture for Enfuse's products, overseeing the infrastructure that powers secure, on-premises and edge AI deployments. She joined Enfuse from Kubia, which she founded to simplify the deployment of machine learning on edge devices.
Over more than a decade, she has built large-scale data platforms, cloud environments, Kubernetes infrastructure, and LLM inference systems across AWS, NVIDIA Jetson, Proxmox, vLLM, Terraform, Ansible, and Rust. She now focuses on production AI infrastructure, observability, and using policy-controlled AI agents to accelerate software development without compromising reliability or security. She is based in Seattle.
Bryan Wilkinson, Ph.D.
Head of Foundational Model LLM & R&D
Bryan is an AI executive and researcher leading the Foundational Model LLM and R&D teams. With a Ph.D. in Computer Science from the University of Maryland, Baltimore County, he brings a deep academic and practical foundation to the development of next-generation large language models.
Throughout his career, Bryan has led data science and machine learning initiatives across diverse sectors, architecting advanced NLP systems, conversational AI platforms, and predictive intelligence models. He specializes in bridging the gap between cutting-edge AI research and scalable enterprise solutions, driving the strategic vision for both AI development and technical enablement.
The bench
The people who deploy inside your boundary
Forward deployed engineers, data scientists, and platform architects across sovereign AI, physical AI perception, and GPU infrastructure. Hover a card to see what they bring.
Engineering lead for warehouse and underwriting systems, shipping pragmatic tools that remove manual work.
Built a fire-door inspection app that auto-generates reports
Who We Are
Enfuse lets you plan with frontier models and execute inside your own boundary — the complete stack for building, deploying and operating AI applications on infrastructure you control, from governed enterprise applications to computer vision, robotics and edge perception.
- Physical AI & Perception
- Sovereign AI Platforms
- AI Infrastructure & GPU Systems
- Forward Deployed Engineering
Most AI firms want your data and play by their rules. We deliver on-prem, air-gapped GenAI for regulated enterprises, government, and defense — on your hardware, under your policies, with zero data leaving.
We don't rent you access. We leave you owning the capability.
How We Work
Services-Led Software — pioneered at Pivotal Labs, refined through hundreds of enterprise AI deployments.
Mission-Ready Pods
Balanced teams of 4–8: UI/UX, Platform, Software, and Product. We own outcomes, not tasks.
Forward Deployed Engineering
Engineers embed with your team. We don't hand over code and walk away — we ensure it works in production.
Knowledge Transfer
Every engagement includes structured pairing and documentation. When we leave, your team owns the capability.
NVIDIA Partnership
Preferred Partner: Blackwell, DGX Spark, NIM, Triton, and Fleet Command.