Seven products · ten app surfaces · all original work
AI products, built end to end.
Voice agents, RAG assistants, workflow automation, OCR search — each one a working application you can open, click, and verify. These case studies show what was built and how.
Fictional demo businesses · no invented clients, metrics, or testimonials
01LumaEvery service enquiry gets an owner.A lead follow-up workflow for service businesses: capture from forms, calls, and WhatsApp into one inbox — with AI reply drafts, owner approval, and an audit trail.React + TypeScriptAI reply draftsApproval workflowRead the case study 
02VoiceLead AIA receptionist that never misses a call.Real-time voice AI that answers, qualifies, and books appointments — while the front desk watches it happen live.Real-time voice (WebRTC)Streaming transcriptionLive qualificationRead the case study 
03ClientPilot AIA form fill becomes a booked appointment. Nobody touched it.AI intake that answers from the business’s own documents, follows up by SMS, and books the calendar — with a visible audit trail for every step.RAG with citationsWebhook automationSMS follow-upRead the case study 
04EduNavigator AIAnswers with receipts: the explanation and the data, in one reply.A study-abroad advisor that grounds explanations in verified sources and derives program tables from a live catalog — with the admin console that keeps both honest.Hybrid RAG + SQLCitation provenanceStructured catalogRead the case study 
05EcoMind AIAsk 200 documents a question. Get the page it came from.An enterprise knowledge assistant where every streamed answer cites its source — and clicking the citation lands on the exact highlighted passage in the original PDF.Enterprise RAGStreaming answersSource viewerRead the case study 
06FlowForgeWatch the workflow run. Watch the spreadsheet fill itself.Marketing automation on a node canvas — triggers, enrichment, email, SMS, and Canva wired together, with runs you can watch execute step by step.Workflow orchestrationNode canvas (SVG)Sheets syncRead the case study 
07ProductSense AIIt read the scanned datasheet. And it can prove it.OCR + vector search + chat over a messy industrial catalog — where the answer quotes a number that exists only inside a photocopied scan.OCR extractionVector searchGrounded chatRead the case study Next step
Want one of these for your business?
Describe the workflow — the written fixed scope, price, and timeline come before any money moves.