I build the software business runs on
Now I do it with AI. Agentic systems that do real work inside your operations — not demos, not pilots that quietly die. Bring me your hardest AI use case and I will make it real. Bring me a product idea and I will help you build something people actually use.
For sixteen years, large companies have handed me the system their business depends on — the one that is too slow, too fragile, or too tangled to fix from the inside. I sit with the people who use it, build the fix myself, and stay until it is running in production.
Making systems that were never meant to talk to each other work as one — legacy platforms, vendors, and modern cloud services, joined without downtime.
From an idea on a whiteboard to something real customers use every day — designed, built, launched, and looked after once it is live.
AI that does actual work inside a business: reading calls, answering questions, handling the repetitive tasks — running safely on real company data.
Websites and web platforms that carry real traffic and real money — fast, secure, and built to keep working as the business grows.
I write the code, own the security review, and stay on the pager. However deep the problem goes, that is the range.
LLM and speech systems that run against live production data under HIPAA and PCI constraints — not demos. Whisper call-centre automation, intent routing, RAG, evaluation harnesses.
1,400+ production APIs across REST, GraphQL federation, gRPC and legacy SOAP bridges — with the gateways, throttling, and zero-trust security that keep them alive at scale.
AWS-native platforms from Terraform up: EKS, Lambda, Redshift, Kinesis, EventBridge. Streaming and CDC pipelines moving tens of millions of records an hour.
Websites, mobile apps, and SaaS platforms built and shipped for startups and mid-market businesses — design, build, deploy, and the operations after launch.
When the software a business runs on starts holding that business back, someone has to go in and fix it without stopping the business. That is the job.
“At dinner rush, orders piled up faster than the system could pass them to the kitchen.”
I rebuilt the ordering platform so it stays fast on the busiest night of the year.
Orders now reach the kitchen a third quicker — in every store, at every rush.
“Twelve thousand members called every day, and nobody could tell you what they were calling about.”
I built an AI assistant that listens to every call, writes the summary, and routes it to the right team.
Staff stopped typing notes and went back to helping people.
“The bank wanted to open its data to partners, but one audit failure would have stopped the programme.”
I designed the secure connections and sat with their compliance and risk teams through every review.
Passed the toughest audit in the industry with zero findings.
“Forty systems disagreed about what was in stock, so shoppers were told the wrong thing.”
I put one dependable layer over all of them and kept eighteen million products in step.
One honest answer about what is actually on the shelf.
“The utility collected enormous amounts of meter data and could not do anything useful with it.”
I turned that firehose into something other companies could safely plug into.
A cost centre became a platform other businesses build on.
“Every project started from scratch, so delivery dates were guesses and quality was luck.”
I introduced an AI-assisted way of working and trained sixty engineers to use it.
Work ships 65% faster, with far fewer problems reaching customers.
Different industries, same pattern: something important was broken, and it had to be fixed while the business kept running.
Systems that carry patient and member data, where a mistake is not just a bug. Built under HIPAA, with PHI encrypted and audited end to end.
Not side projects — live products with real customers, built and run end to end.
E-commerce
Indian grocery, online
An online Indian grocery store specialising in pulses, staples and household goods — catalogue, cart, payments and fulfilment built for a category with thin margins and heavy SKU counts.
Visit →
Real Estate / PropTech
Live MLS search for the Phoenix metro
A real-estate platform for buyers and sellers across Scottsdale, Gilbert, Chandler and the wider Phoenix metro — live IDX/MLS property search with real-time data, advanced filtering, and lead capture wired into the brokerage’s workflow.
Visit →HealthTech / AI
Pharma Insights AI
An AI-powered intelligence platform delivering analytical capability to the pharmaceutical industry — built for professionals who need defensible, data-driven insight in a heavily regulated market.
Visit →
FinTech
Cross-border family wealth platform
A wealth platform for families managing money across multiple countries — consolidating accounts, currencies and jurisdictions into a single view.
Visit →FinTech / SaaS
Financial infrastructure tooling
A financial stack product in active development, applying the same spec-first, API-first engineering discipline used across enterprise engagements.
Visit →Consumer / Community
A movement, not just a brand
Screen-free play for children — physical activities, collectibles and games placed in high-traffic family locations like malls and restaurants. Includes a community idea feed with 200K+ contributions.
Visit →
E-commerce
Personalized return gifts
An e-commerce platform for customized return gifts serving the Arizona market — sister brand to NoGadgets, covering party and event gifting end to end.
Visit →Marketplace
On-demand water tanker booking
A booking marketplace connecting households and businesses to water tanker suppliers — scheduling, dispatch and delivery tracking for an entirely offline industry.
Visit →Client Work / Healthcare
Digestive disease specialists, Phoenix
Web presence for a multi-location gastroenterology practice across Glendale and Goodyear, Arizona — provider profiles, services, and appointment scheduling.
Visit →Deployed into 18 enterprise customers across fintech, healthcare, retail, energy and Web3. Own discovery through build, production and handover. Lead cross-functional teams while staying hands-on in architecture and code.
20+ AI applications shipped into live customer operations — LLM, speech and agentic workflows built against real production data under HIPAA and PCI controls.
Built the public API platform at InsideView including a two-tier throttling engine and usage, quota and cost-metering framework. At CA Technologies, developed, tested and supported CA Wiki and built automation frameworks for CA ITCM and CA Spectrum.
Tell me what you're building and where it's stuck. I read everything that comes through here.