The entire AI environment is governed from one place rather than defended in pieces.
Chakali / AboutSecurity cannot be added
Security cannot be added
around AI.
It has to be engineered into how the AI operates. That conviction — formed over more than a decade of building advanced cybersecurity systems and securing air-gapped environments — is why Chakali exists.
01Why we built Chakali
A gap opened, and
nothing was closing it.
With more than a decade of experience building advanced cybersecurity systems and securing air-gapped environments, we recognized a critical emerging challenge: AI models and autonomous Agents have evolved beyond the boundaries of traditional security controls.
At the same time, organizations using frontier cloud AI are achieving capabilities far beyond previous expectations — while governments and enterprises that cannot use public-cloud services risk falling behind.
Chakali was built to close this gap. It is a secure-enclave AI operating system and agentic harness that enables organizations to deploy, secure and continuously evolve advanced AI within their own on-premises or air-gapped environments.
02What makes Chakali different
Convergence, not another
point solution.
Chakali's key differentiator is the convergence of advanced AI capabilities and deterministic security controls within one secure-enclave, on-premises platform.
Unified in one platform
- AI Agents
- Multi-agent collaboration
- Workflows
- Secure Skills
- Organizational knowledge
- Memory
- Multi-LLM reasoning
Separate AI systems, each with its own access and its own blind spots, stop accumulating.
Agents, knowledge and workflows continuously build upon one another, so the organization gets better rather than busier.
New models, Agents, tools and technologies can be introduced securely as the AI landscape evolves.
03The three questions Chakali addresses
Every deployment comes back
to these three.
- 01
How can organizations adopt advanced agentic AI without compromising security, governance or sovereignty?
- 02
How can they continuously evolve as AI models, Agents and tools advance?
- 03
How can they control what AI is permitted to access, decide and execute?
04Who built it
The person behind
the architecture.
Hud Seidu DaannaaFounder & CTO, RabitaNoorCreator of Chakali AI OSFull profile
Hud Seidu Daannaa is a cybersecurity architect, AI security practitioner, educator and technology founder focused on building secure AI systems for enterprise, government and highly regulated environments.
His experience shaped the philosophy behind Chakali: as AI systems become increasingly autonomous and gain access to organizational data, tools and infrastructure, security cannot simply be added around AI — it must be engineered into how the AI operates.
He created Chakali AI OS as a secure-enclave AI operating environment that lets organizations build and operate AI Agents, workflows, knowledge systems and intelligent applications while maintaining control over permissions, data, tools, execution and human authority. A major focus of his work is bringing advanced AI capabilities into on-premises, sovereign and air-gapped environments.
Background
Education and community
- Cybersecurity lecturerMiddlesex University Dubai
- Wazuh AmbassadorWazuh
- Subject Matter Expert, SecAI+CompTIA
- Global AmbassadorUniversity of Surrey
Research interests
The guiding principle
Intent is not authority.
An AI system may understand what a user wants. That should never automatically give it the authority to access data, execute commands or change critical systems.
Everything in Chakali — the permission model, the policy engine, the approval gates, the evidence record — is an expression of that one sentence.
Work with us