AQUNAMA Technical Foundation — Architecture, Security, Deployment

Our engineering approach

We don't pitch architecture diagrams in a sales meeting and then build something different. We don't rip and replace your existing stack. And we don't sell platforms that lock you in.

Every AQUNAMA deployment follows the same engineering principles:

• —Integrate with what you already have — your ERP, CRM, accounting, telephony, document storage
• —Use modern, enterprise-proven components — no exotic tech we can't support five years from now
• —Senior engineering and AI-accelerated development — fewer people, faster delivery, higher quality
• —Fixed-scope, fixed-price proposals — you know what you're getting before work begins
• —Cloud-agnostic deployment — AWS, Azure, GCP, on-prem, or our own ISO 27001-certified data center
• —Honest commercial terms — ownership, access, and licensing agreed clearly upfront, per engagement

The AQUNAMA tech stack

Layer / What we use

Security and compliance

Security is part of the architecture, not a layer added afterward.

We work with clients in regulated industries (insurance, financial services, healthcare) and align deployments with their compliance requirements — including NIS2 readiness, sector-specific frameworks, and internal audit standards.

• —Security-certified hosting — our own ISO 27001-certified Tier III data center in Prague, plus major cloud provider regions (AWS, Azure, GCP)
• —Data encryption in transit (TLS) and, where applicable, at rest
• —Access controls based on the principle of least privilege
• —Multi-factor authentication for administrative access
• —Audit logging of access to sensitive data and operations
• —Backup and disaster recovery with documented RPO/RTO targets per engagement
• —Regular vulnerability management and dependency monitoring
• —GDPR-compliant by design — including SCCs for any non-EEA data transfers

Data handling and sovereignty

Data sovereignty is a per-project decision. We deploy across EU regions of major cloud providers (AWS, Azure, GCP), our own ISO 27001-certified data center in Prague, or your own infrastructure — based on your compliance, performance, and cost requirements.

Whichever model is chosen, every client gets:

• —Documented data flows — you know exactly where your data lives and moves
• —GDPR-compliant defaults — including data residency, retention controls, and access logging
• —Transparent sub-processor disclosure

Deployment options — cloud, on-premises, hybrid

Public Cloud (AWS, Azure, GCP, and others): Deployment on major cloud platforms — typically chosen when clients have existing cloud infrastructure, regional requirements, or want full elasticity.

AQUNAMA Data Center: Hosted in our own Tier III, ISO 27001-certified data center in Prague. Strong fit when clients want EU data sovereignty without managing their own cloud account.

On-premises: For clients with strict regulatory requirements, air-gapped environments, or sovereignty needs beyond the EU. Including GPU-equipped servers where AI inference performance requires it.

Hybrid: Some components on a public cloud or our DC, some on yours. Common in financial services, healthcare, and other regulated industries.

Integration approach

Most AQUNAMA deployments connect to systems you already use:

• —ERPs — SAP, Microsoft Dynamics, Odoo, and custom systems
• —CRMs — Salesforce, HubSpot, Pipedrive, custom CRMs
• —Accounting — major Czech and international accounting platforms
• —Telephony — Twilio, Vonage, Genesys, and PBX systems
• —Document storage — SharePoint, Google Drive, Dropbox, custom systems
• —Messaging — WhatsApp Business API, SMS providers, email infrastructure
• —Calendar & scheduling — Calendly, Google Calendar, Microsoft 365

The AI layer

We use AI in two distinct ways: inside client deployments and inside our own engineering process.

Inside our engineering process: We run an integrated layer of AI agents across Claude, ChatGPT, Hermes, and other platforms to accelerate development, generate test cases, document architecture, and compress timelines.

Inside deployments:

• —Custom voice agents (STT + LLM + TTS) for inbound and outbound call automation
• —AI Custom Agents — chatbots, internal assistants, agentic workflows
• —Multilingual speech analysis for QA and sentiment monitoring
• —Document classification, extraction, and processing pipelines
• —Workflow orchestration that combines deterministic logic with LLM-driven decisions

AQUNAMA's deployment model

Phase / What happens

Commercial models and ownership

Different projects warrant different commercial models:

• —Custom one-off builds — typically with full source code in your repository and no platform lock-in
• —AQUNAMA proprietary platform deployments — where we use our own reusable platform, customized to your operations and hosted by us
• —Hybrid engagements — combining custom development with AQUNAMA components, with clearly defined IP and access boundaries