A Delaware Corporation built around AI.
Gilligan Tech Inc. was incorporated as a Delaware Corporation in October 2024 by Dave Gilligan — a full-stack developer and AI system architect with over a decade of experience building production software for businesses across Europe and North America.
The company was founded on a straightforward observation: the gap between what enterprise AI teams can deploy and what a 10-person business can access is enormous — and almost entirely artificial. The models are available. The infrastructure is available. What's missing is a focused company that builds the integration layer cleanly, ships it as a product, and prices it for real businesses.
We build on Google Vertex AI, AWS Bedrock, and Azure OpenAI — the same infrastructure the Fortune 500 uses — and deliver it as turn-key SaaS products and managed implementation for companies that don't have a machine learning team on staff.
Three products. One platform.
LLM Integration Platform
Multi-model orchestration across Vertex AI, Bedrock, and Azure OpenAI. One API that routes each request to the optimal model — by capability, latency, and cost — with no vendor lock-in. Drop Claude, Gemini, or GPT-4o into your product in days, not months.
Document Intelligence & RAG
Upload any document — PDF, Word, spreadsheet, contract. Ask any question in plain English and get a grounded, cited answer powered by retrieval-augmented generation. Deploy in hours. Works on your documents, with your access controls, in your cloud region.
AI Workflow Automation
LLM-powered triggers, data extraction, classification, summarisation, and routing — connected to the tools your team already uses. No-code for the standard library. Full API access when you need it.
Company facts
Why small business?
Large enterprises have ML teams, data science departments, and vendor relationships that cost millions. A 20-person accounting firm, a regional law office, or a growing SaaS startup has none of that — but faces the same volume of documents, support tickets, and repetitive workflows. We exist to close that gap.
Four commitments we keep on every project.
Your data never trains a model
Customer documents, queries, and outputs are processed via inference-only API calls on all four cloud platforms. Nothing is used for fine-tuning or training. Period.
No vendor lock-in
Our routing layer works across Google, AWS, Azure, Cloudflare, NVIDIA, and eight additional providers. Switching models is a configuration change, not a rewrite.
Full auditability
Every model call is logged with the model used, token count, latency, and response. Available for download at any time. No black boxes.
US data residency by default
All data is processed and stored within US-based cloud regions. EU residency is available on Enterprise plans via Azure and Mistral AI hosted infrastructure.
See the platform in action.
Book a 30-minute demo and we'll show you the exact workflow for your industry — live, with your data if you have it.