Independent products, a client AI enhancement and the systems supporting the work. Start with a decision: reviewed agent actions in nsso, a human concierge handoff in 24Seven, or voice capture and recovery in Dreamsea.
Independent product
People bringing their profile, work and products into one public identity.
Client AI enhancement
A boutique travel business whose clients preferred conversation to catalogue browsing.
Independent product
People trying to capture a dream before the details disappear.
Qadam is a guarded paper research experiment; RazinFlix brings curation and automation together. The workspace and tools have their own implementation and evidence boundaries. “Selfware” is my shorthand for useful products I build for myself and others.
Independent product
Me and trading collaborators reviewing financial research and paper-only decisions.
Independent product
Me, family and friends choosing films from recommendations we trust.
Supporting system
Product managers using AI coding assistants for recurring product work.
Supporting system
People turning saved links and screenshots into knowledge they can review and reuse.
Supporting system
Builders who need to inspect model usage and cost before choosing where to optimise.
Supporting system
AI-product builders needing attributable context from separate source collections.
Shopify, but the product is you: one owned identity surface where profile, proof, projects, products, links, and Deity AI turn self-presentation into self-clarification.
A paper-only research and trading system that compares data sources, records strategy reasoning, and applies guarded rules before simulated orders.
A voice-first dream journal with multimodal generation and philosophy-specific interpretation.
A catalog-grounded travel concierge that plans across Shopify inventory and hands off to a human agent.
A streaming-style film library built from trusted human taste, enriched by metadata scripts, trailers, posters, category logic, and manual curation.
A local-first operating system that lets product managers run AI-native workflows through Codex, Claude Code, Cursor, or Antigravity.
A multimodal extraction pipeline that turns social URLs, screenshots, transcripts, captions, and carousel posts into structured knowledge.
A standalone Python library for metering LLM usage, tracking spend and budget alerts, and routing with cached eval scores plus explicit fallback rules.
A zero-dependency Python RAG layer for ingesting, chunking, embedding, storing, retrieving, and packing cited context across product namespaces.
The projects use different implementation stacks; the five-layer framework helps compare their choices.
My product method · AI-Native Product OS
AI can make a first working version cheaper to build. It also creates outputs that need to be checked. I use five areas of responsibility and a repeatable development loop to connect the work to a customer problem, clear decision rights and evidence of what happened.
Governance and human responsibility apply throughout. The checks and release gates depend on the consequences of getting the task wrong.
The framework is this way of thinking. The workspace is the installable GitHub toolkit. The courses teach it on Udemy and Coursera.