Straight Answers.
No Buzzwords.
Everything you’ve
wondered about private AI—
spelled out clearly.
Ask us anything—
except for your data. That stays with you.
01
General Overview
Everything you need to know before exploring the details.
FortiqAI is a privacy-first AI platform that runs on your infrastructure — on-premises or inside a private cloud — so sensitive documents never leave your control. It combines a user-friendly React web UI with a Node.js backend and local LLMs/vector stores (e.g., Qdrant) to provide search, summarization, Q&A, and document analytics.
Think: the power of modern LLM features with enterprise-grade data control and the ability to extend or customize behavior to your processes.
FortiqAI is built for organizations that must protect proprietary or regulated information: law firms, pharmaceutical R&D labs, financial institutions, healthcare providers, government teams, and any business with critical IP or compliance needs. It suits teams of all sizes — from small legal boutiques that need faster document search to large enterprises requiring custom integrations and strict audits.
Cloud services send your documents to external servers for processing; FortiqAI keeps everything inside your environment.
That brings three practical benefits:
(1) compliance — you control data residency and access;
(2) predictability — no external data transfer or per-request cloud costs;
(3) customization — you can adapt models, connectors, and UI to your exact workflows.
The trade-off is you manage infrastructure and updates, but we provide tooling and options to make that straightforward.
02
Customization & Services
Your workflows, your rules — product + professional services.
Yes — FortiqAI is modular. Typical customizations include: custom connectors (to case management, DMS, or internal DBs), UI changes (dashboards, role-specific views), automated workflows (document tagging, alerts, email routing), and custom extraction or entity types (legal clauses, trial data, drug names).
We work from a short discovery to map your processes, then deliver a scoped plan (prototype → build → test → handover).
Absolutely. If your needs go beyond adapting the product, we build custom private-AI systems from scratch (architecture, data pipelines, model fine-tuning, UI, and integrations).
Typical engagement: discovery workshop → solution design with milestones → prototype/PoC → full implementation → operational handover and training. You keep ownership of your data and IP; we can operate under NDAs and bespoke SLAs.
Small changes (UI tweaks, single connector): ~1–3 weeks.
Medium scope (new integrations + modest UX + rule engine): ~3–8 weeks.
Large/bespoke (full custom platform, model fine-tuning, heavy integrations): 2–4+ months.
We usually start with a short discovery (1–3 days) and deliver a fixed-scope proposal or time-and-materials plan. Pricing depends on scope, required security/compliance, and whether we provide on-site work or white-glove deployment.
02
Deployment & Technical Details
From hardware to updates and integrations — how it runs and stays secure.
Deployments commonly use containerized packages (Docker / Helm charts) or VM appliances.
Steps:
(1) discovery & environment review,
(2) package delivery and installation,
(3) ingest/connect documents (manual upload or connector),
(4) indexing and initial model setup,
(5) validation and user training,
(6) go-live.
We can run the deployment remotely, assist on-site, or hand off to your ops team with full docs.
Sizing depends on dataset size and the model you run.
Typical guidance: a modern multi-core CPU, 32 GB RAM minimum for small-to-medium workloads, NVMe SSD storage for fast indexing, and a GPU (NVIDIA T4/A10/A100 class) to accelerate inference for larger LLMs.
For enterprise loads or many concurrent users, we recommend multi-node setup and dedicated GPUs. We provide a sizing checklist during discovery to produce an exact spec for your case.
No — FortiqAI can run fully offline once installed. Internet is only needed if you choose to download updates from us or to enable optional external integrations. For air-gapped environments we deliver signed update bundles and documentation for manual application so connectivity is not required.
We provide signed update packages and a documented update process. Options: manual patching (air-gapped), scheduled secure update windows, or one-time secure connection for patching under customer control. We advise standard hardening: least-privilege accounts, encrypted disks, TLS for internal services, and routine vulnerability scans. We also provide guidance for SIEM/audit log integration.
Common integrations: PostgreSQL, Qdrant (vector store), SharePoint / network shares, S3-compatible storage, internal DMS, and REST/GraphQL APIs. We also support ingestion of standard files — PDF, DOCX, PPTX, CSV, images (OCR) — with configurable parsing & chunking. If you have a proprietary system, we build a connector as a deliverable.
All data — original documents, indexes, and vector embeddings — remain on your infrastructure. FortiqAI supports encryption at rest (your keys or KMIP integration), TLS for in-transit protection, and role-based access control (RBAC). You decide user privileges; we can integrate with enterprise SSO (SAML/LDAP/OIDC) and produce audit logs so every query or export is traceable.
02
Pricing & Support
Clear options and dependable help for production use.
Support tiers range from standard email support and update bundles to enterprise SLAs with 24/7 coverage and a designated support engineer. We provide training: admin training for ops teams, user training sessions, and onboarding materials (runbooks, install guides). Custom support (on-site or dedicated hours) is also available as part of enterprise engagements.
Pricing generally has a few components:
(A) product license / instance fee,
(B) customization development (one-time) or subscription for managed/custom work,
(C) optional support & maintenance (annual), and
(D) infrastructure costs (your servers/GPU).
For transparent procurement we offer: discovery + fixed-scope quote for one-off projects, or a modular plan (license + annual support + hourly dev retainer). We’ll prepare a tailored commercial proposal after a short discovery call.
Typical start steps:
Contact & NDA (if required) — schedule a short discovery call.
Discovery workshop — map objectives, data sources, security constraints.
Proposal & SOW — scope, timeline, cost, and acceptance criteria.
Proof-of-Concept (optional) — pilot on a subset of data (2–4 weeks).
Deployment & training — rollout, user onboarding, operational handover.
Support & iteration — ongoing tweaks, custom features, or expansion.
We’ll provide a checklist for each phase so you know who does what and when.