Enterprise AI solutions for business
A production-ready enterprise platform: secure deployment, integration into your stack, and transparent service terms.
Standardized rollout, tailored to each customer.
Mini
Sales & customer service
$59
per month
A top-tier solution for small business. A real boost to sales and service.
- 3–5 closed tickets per day
- 3–5 new leads per day
- Top-tier LLM models
- Website integration
- Product sync (unlimited)
- Gmail / Calendar sync
- Analytics & reports
- 24/7 technical support
Choose Plan →
Basic
Sales | customer service | API
$269
per month
The full range of AI integrations for mid-size business. Easily replaces 3–4 employees.
- 15–25 closed tickets per day
- 15–25 new leads per day
- Top-tier LLM models
- Website integration
- WhatsApp number
- WhatsApp AI agent
- Telegram AI agent
- Product sync (unlimited)
- Gmail / Calendar sync
- Monday sync
- Priority sync
- In-chat payments (Cardcom)
- Analytics & reports
- 24/7 technical support
Choose Plan →
Recommended
Standard
Sales | customer service | API | BPM
$549
per month
A fully equipped premium AI machine. Maximum efficiency for mid-size businesses.
- 50–100 closed tickets per day
- 50–100 new leads per day
- Website integration
- Business mobile / WhatsApp number
- WhatsApp AI agent
- Telegram AI agent
- Outbound AI voice calls
- Inbound AI voice calls
- Product sync (unlimited)
- Gmail / Calendar sync
- Monday sync
- Priority sync
- In-chat payments (Cardcom)
- CRM API
- Rivhit integration
- Analytics & reports
- Conversational Audit
- Lost Lead / Revenue Leak Audit
- Response Time Audit
- Top-tier LLM models
- 24/7 technical support
Choose Plan →
Enterprise
AI Microservices full bundle
$1,890
per month
All WAAI systems and AI Microservices pushed to their limits. The absolute standard of quality.
- 800–1000 closed tickets per day
- 800–1000 new leads per day
- Business process integration (BPM)
- Outbound AI voice calls
- Inbound AI voice calls
- Website integration
- WhatsApp number
- WhatsApp AI agent
- Telegram AI agent
- Product sync (unlimited)
- Gmail / Calendar sync
- Monday sync
- Priority sync
- In-chat payments (Cardcom)
- CRM API
- Rivhit integration
- Analytics & reports
- Monthly live consulting
- Full AIM API access
- Conversational Audit
- Lost Lead / Revenue Leak Audit
- Response Time Audit
- Top-tier LLM models
- 24/7 technical support
Contact us →
Frequently Asked Questions
Where is data physically stored? Is data residency supported?
Contractual data residency: EU (Frankfurt / Amsterdam), Israel (Tel Aviv), and optionally US-East (Virginia). All customer data — conversations, embeddings, logs, operational snapshots, backups — never leaves the boundaries of the selected region.
For SaaS deployments we use AWS / Azure with account-level residency pinning. For VPC and on-prem deployments, data stays entirely inside your infrastructure. Full sub-processor list and Data Processing Addendum (DPA, GDPR Art. 28) are provided before MSA execution.
How are data isolation and BYOK implemented?
Single-tenant architecture: each customer gets an isolated VPC with dedicated databases, message queues, and runtime. No shared compute or shared storage between tenants — ever.
BYOK is supported via AWS KMS, Azure Key Vault, and GCP KMS — master keys remain in your cloud account. WAAI operates through cross-account roles with encrypt/decrypt grants only, never holding the keys themselves. AES-256 encryption at rest for all persistent data, TLS 1.3 for all network traffic. Key rotation policy, incident response procedure, and architectural diagrams are included in the DD pack.
What deployment models are available?
Three options:
SaaS Managed — WAAI-managed cloud, fastest time-to-value (4–6 weeks), best fit for non-regulated workloads.
VPC Dedicated — isolated tenant inside your AWS / Azure / GCP account, integrated with your IAM / SSO / SAML / VPN, full network policy control on your side.
Full On-Premise — Kubernetes cluster in your data center, optionally air-gapped (no outbound internet), with support for self-hosted LLMs (Llama, Mistral, custom).
All three options run on a single codebase — your choice does not constrain platform functionality.
What does the SLA include and what are the service credits?
Contractual SLA: 99.95% monthly uptime, p99 latency < 800 ms, RTO 4 hours / RPO 15 minutes.
Service credits on breach: 10% of monthly fees if uptime drops below 99.95%, 25% below 99.5%, 50% below 99.0%. Credits are applied automatically in the next billing cycle — no ticket required.
Public status page with 24-month uptime history, post-mortem incident reports within 24 hours, and PagerDuty / Opsgenie integration for real-time alerting on the customer side.
Is our data used to train shared models?
No. Conversation content, documents, knowledge bases, and customer metadata are never used to train foundation models or any model that other customers see. This is contractually enforced in the MSA and DPA.
Fine-tuning is performed only on your data and only inside your dedicated tenant. All weights of fine-tuned models belong to the customer and can be exported in standard formats (GGUF, safetensors) at contract termination.
WAAI maintains zero-data-retention agreements with upstream LLM providers (OpenAI Enterprise, Anthropic, Azure OpenAI) — inference requests and responses are not stored at the provider.
How are hallucinations and answer quality controlled? Is there human-in-the-loop?
A multi-layer governance stack:
1. Constrained generation — RAG verification of every factual claim against your knowledge base; unsourced answers are blocked.
2. Confidence-score gating — low model confidence automatically escalates the conversation to a human operator via the HITL queue.
3. Topic boundaries — out-of-scope requests (legal advice, medical diagnoses, etc.) are blocked at policy level and routed to HITL.
4. Full audit log — every AI agent reply is stored with the full trace: prompt used, RAG sources, confidence score, model version.
Aligned with ISO/IEC 42001 (AI Management Systems) and EU AI Act 2026 requirements for high-risk use cases.
What happens at contract termination? Vendor lock-in?
Exit policy is contractually defined in the MSA: a 90-day transition window post-termination, during which WAAI commits to a technical handover to your in-house team or to another vendor.
Full data export in open formats: conversations and metadata as JSON / Parquet, fine-tuned model weights as GGUF / safetensors, agent configurations and prompt templates as YAML. No proprietary formats. No encrypted-only exports.
All integrations are built on open APIs (OpenAPI 3.1 / AsyncAPI 2.6) — after migration you can plug any other AI platform into the same endpoints without rewriting client code.
.png){kind=logo}
