Work & products

AI products we run and systems we have delivered

Two AI products built and operated by Dassjay Labs, plus the kinds of platforms we build for clients. Client names are kept confidential.

Agentic document intelligence for regulated teams

Akurion.ai

Akurion reads your files, plans its own searches and cites the page behind every statement, so regulated teams and the agents they build get governed answers they can verify.

  • Agent mode that plans multi-step searches across a project's documents and the web
  • Graph RAG: entities and relationships extracted so answers can follow links between documents
  • Page-level citations on every statement, with answer-quality evaluation built in
  • Vector store management, enrichment and memory across conversations
Why RAG demos break on real documents, and how Akurion handles itNarrated explainer that starts with why RAG prototypes fail on large, changing document sets, then shows Akurion product screens: connectors, auto-generated metadata keys, chat filters, the Graph RAG relationship map, cited source pages opened in chat and the MCP playground.

Documents shown in the Akurion videos are fictional sample data.

SEC filings intelligence for investors

The Filing Fool

Every insider filing, congressional trade, fund position change and federal contract on one page per company, with each dollar sized against the company's own revenue or market cap.

  • Continuous ingestion pipelines over SEC EDGAR, congressional disclosures, lobbying records and federal contract awards
  • Entity resolution linking insiders, funds, members of Congress and companies across sources
  • A per-ticker timeline that correlates every event with price and sizes it against revenue or market cap
  • Signals ranked by strength: fund conviction, cross-source agreement, unusual insider clusters and industry pulse
Insider Scanner: open-market insider buying and sellingNarrated walkthrough of the Insider Scanner, which ranks open-market buys and sells by officers and directors by dollar value, filtered by company size and time window. Each trade links to its Form 4, and pre-planned 10b5-1 trades are left out.

The Filing Fool is not investment advice. Every figure comes from public filings.

Selected engagements

Platforms and systems we have built

A cross-section of client and internal work, across AI automation, data and full-stack product delivery.

Agent and client platform for a life insurer

A two-audience platform with plan explainers, self-serve calculators, a plan recommender, a lead pipeline, need analysis and talk tracks for agents.

Web platformSales enablementInsurance

AI marketing platform for real-estate developers

Upload a project brochure and the platform reads it page by page, builds a searchable knowledge base, enriches prospects and drafts campaigns.

RAGMarketing automationReal estate

Local sanctions-intelligence engine

Resolves aliases, transliterations and shell entities from public watchlists into one graph, with a local open-weight model making the judgment calls.

Entity resolutionGraphPrivate LLM

Sanctions screening platform

Search and screening across OFAC, UN, EU and UK lists with entity-type and country filters for compliance teams.

ComplianceSearch

Verified local business registry

A Waterloo Region directory of home-service businesses, enriched with licensing and compliance data from authoritative sources.

Data enrichmentDirectoryOntario

Lead-generation marketplace

A multi-step request flow that matches homeowners with qualified builders, with quoting and pay-per-lead billing.

MarketplaceWeb app

Local voice and narration tooling

On-device text-to-speech and consent-based voice cloning on Apple Silicon. Nothing leaves the machine.

Speech AIMLXPrivate

Automated product-film pipeline

Browser-recorded walkthroughs and animated explainers with AI narration, music, captions and social cuts, defined in one config file.

Video automationPlaywright

Have a product or platform in mind?

A 30-minute call is enough to tell whether AI is a fit, what a first project would look like and what it would take to run it in production.