I build AIsystems thatoperate.
Senior full-stack engineer and founder with 25+ years shipping software, commerce platforms, AI products and automation. Current work spans agentic workflows, retrieval, tool use, evaluations, guardrails and observability.
I turn complex ideas into software people can use and teams can operate, then stay with the system long enough to make it reliable.
What I build
AI agents
Autonomous workflows that research, decide and act inside explicit controls.
Data pipelines
Pipelines that show where information came from, how current it is and how it changes.
Automation
Reliable jobs and tools that take repetitive work off people.
Platforms
Web products built for everyday use, reliability and growth.
Integrations
APIs and services connected through narrow, observable boundaries.
Outcomes
Software that solves a defined problem and can be measured, maintained and improved.
Selectedwork
Products Jason owns end to end: designed, built, shipped and operated.
Viral Worx
An AI growth workspace for creators on YouTube, TikTok, Instagram and Facebook: analytics, scheduling, bulk AI editing, media generation and coaching in one place.
The AI Bookie
Sports decision support for adults 21+: grades today's sportsbook markets from the consensus of the books, and shows every reason, every source and how fresh the data is.
AI Paper Trader
A practice-first trading workspace for U.S. stocks, ETFs and crypto: research, strategy review, order previews, approval gates, risk controls and audit logs, all on paper.
Your Chess Career
A free tool for Chess.com and Lichess players to explore their whole career (ratings, openings, opponents, tournaments) and review their own games with an in-browser engine.
Technologywith a job.
From interface to infrastructure. Every tool earns its place in the system.
Build the experience
Fast, responsive web products with clear service boundaries and reliable data flow.
Make work move
Retrieval-grounded and agentic workflows built with narrow tools, structured outputs, explicit approvals and measurable operating controls.
Turn intent into action
Storefronts, mobile commerce, checkout flows, integrations and conversion-focused delivery.
Build to be found
Technical SEO and answer-engine foundations that make product meaning explicit to people and machines.
Built forthe real world.
Products, platforms, teams and operations. Proof beyond the prototype.
DTLR.com + Mobile
Development and optimization across DTLR's web storefront and its iOS and Android commerce app: performance, checkout, analytics, structured data and retail operations.
Art Brand Studios
Led the ThomasKinkade.com rebuild and delivered commerce work across the Art Brand portfolio, including Shopify Plus and digital-collectible integrations.
Nestlé Health Science
Coordinated business and technical teams across a multi-brand consumer-health commerce program.
Daniel Markus Inc. / Pierce Bainbridge
Rapid commerce, inventory and legal-platform work across several client properties.
The machines changed.The instinct didn't.
Early computing
Curiosity about how machines work, long before it was a career.
Desktop software
Software you installed, owned and lived inside.
The web
1999: B2B web applications with an international team at Vistaar Technologies.
Web applications
2002–2005: more than 100 client sites at 1800Communications, from one launch a month to three a week.
Data + ad tech
2005–2013: AcxFusion real-time ad reporting and a newsletter audience past one million at Applied Cognetics.
Operations
2013–2020: running growth, finance and technology operations, up to a $100M, 150-person logistics business.
Commerce at scale
2020–now: Shopify Plus, Magento 2 and mobile commerce for Nestlé Health Science, Art Brand Studios and DTLR.
Generative AI
2025: Viral Worx, an AI creator platform, built and run end to end.
Agents
2025–2026: systems that research, decide and act inside explicit controls: the paper trader, a self-healing creator platform, a self-updating family tree.
Next
Currently compiling…
The lab
Not every idea needs to become a company. Some exist because I wanted to know if they could work.
Autonomous Sports Signal Engine
Can market movement become a useful time-series intelligence signal?
EXPERIMENT 002ActivePersistent AI Workflows
How much useful work can software complete when context, tools and state persist?
EXPERIMENT 003ActiveMachine-Operated Systems
What changes when software stops waiting for a human click?
What I'm thinkingabout now.
Agency needs control
An AI system needs clear limits, evaluation, an explanation of its actions and a reliable way to recover from errors.
AI agents
Systems that coordinate work instead of only answering.
Autonomous software
Software designed to operate continuously and safely.
Human + AI development
What happens when the development process itself becomes collaborative.
Robotics
When software intelligence gets a body and a physical environment.
How I build
Practical systems work. Evidence before theatre.
Understand the system
Map the real inputs, outputs, dependencies and constraints before changing anything.
Find the failure points
Look for the boundaries where state, integrations, assumptions or people can break the flow.
Build the smallest working version
Prove the core behaviour before expanding the surface area.
Run it against reality
Real usage exposes what diagrams and optimistic assumptions miss.
Observe what breaks
Logs, state and failure evidence turn surprises into engineering input.
Automate what repeats
Move reliable repeated work out of human memory and into explicit systems.
Keep improving
Ship, measure, learn and revise without pretending the first version is the last.
Patterns inevery system
The same habits show up across everything Jason ships. Each one is traceable to a real project.
Humans approve the risky parts
Automation runs the safe work; anything consequential waits for a person.
Viral Worx · AI Paper Trader · The AI Bookie · Martocci Family Heritage · Stack or Sink
Safe by default
Features ship switched off or locked until someone deliberately turns them on.
The AI Bookie · AI Paper Trader · Viral Worx · Your Chess Career
Freshness is shown, not assumed
Every number says where it came from and how old it is; stale data is labeled or refused.
The AI Bookie · AI Paper Trader · Your Chess Career · Viral Worx
Unknown is not zero
Missing evidence stays missing instead of quietly becoming a number.
Append-only records
Decisions, fixes and imports go into ledgers that can be audited and undone.
Viral Worx · AI Paper Trader · The AI Bookie · Martocci Family Heritage · Stack or Sink
Releases match the commit
What is live is checked against exactly what was tested.
Viral Worx · Your Chess Career · Stack or Sink · The AI Bookie
Cost has a ceiling
Quotas, budgets and scale-to-zero so a good day never becomes a bad bill.
The AI Bookie · Viral Worx · AI Paper Trader · Martocci Family Heritage
AI stays bounded and grounded
Models advise inside deterministic checks and answer from real data, not imagination.
Building softwarewas never reallyabout the code.
Technology has changed dramatically across Jason's career, from early B2B web applications to commerce platforms to AI agents.
What has stayed constant is the process: seeing something that doesn't exist yet, understanding how it should work, and building the system that makes it real. Then staying with it until it is reliable.
More about Jason →