Henry Mayo Asheville, NC

Solutions architect building software since 2013

Software that fits the way people actually work.

Most software problems start somewhere else: a broken handoff, a spreadsheet carrying too much, or a “five-minute” task repeated all day. I work with the people closest to the problem, map the real workflow, and build the fix from end to end. Web, mobile, data, AI—whatever makes sense.

See how I work

01

Start with the real workflow

Before I decide what to build, I sit with the people doing the work. I want to see where the process slows down, where trust breaks, and what would actually help.

Workflow appsOperations toolsPricing systemsDemand analyticsForecasting
02

Make AI earn its keep

Retrieval, fine-tuning, agents, vision—I use AI where it creates a real advantage, and ordinary code where it is faster, cheaper, and easier to trust.

RAGFine-tuningAgentsMCPComputer visionOn-device ML
03

Own the feature end to end

I work across the whole path: interface, APIs, queues, data models, and deployment. Keeping the work together means fewer handoffs, less lost context, and a product that holds up at the edges.

TypeScriptPythonSwiftReactRemixNext.jsShopifyNode.js
04

Keep the machinery boring

Event pipelines, caches, ranking systems, and cloud infrastructure should be reliable, observable, and easy to operate. The product can be interesting. The plumbing should not be.

PostgreSQLFirestoreRedisWebhooksQueuesAWSGCPObservability
05

Build trust in from the start

Sensitive health, education, customer, and payment data need clear boundaries. I design permissions, audit trails, and secure defaults into the system from day one.

HIPAAFERPASOC 2Secure authenticationPermissionsAudit trails
06

Turn ambiguity into a plan

I turn a grower’s instinct into a spec, a technical constraint into a clear tradeoff, and a crowded backlog into the next useful thing to ship. This is the part I like most.

DiscoverySystems thinkingPrototypingMentoringPlain English
Collector platform

Garden

Garden picks up where checkout ends. Customers can sync new orders, opt in to import past purchases, add plants of their own, track collection gaps by cultivar and loyalty progress, and share selected parts of their collection.

Restock alerts only fire when inventory actually returns. Under the hood: verified, idempotent Shopify webhooks, linked identities, Firestore, image processing, caches, and repairable imports.

Order syncCultivar-level trackingPermissioned sharingImage pipelineRestock alerts
Operations system

Garden Admin

Garden Admin is the operating console behind the customer experience. Staff can tune recommendations, track demand velocity, and manage collections and rewards.

A single customer view ties together orders, plants, loyalty, alerts, and linked identities, while a campaign studio shows live send progress. Events are deduplicated and time-decayed, writes are rate-limited, summaries are cached, and every staff action is recorded.

Demand analyticsRecommendation tuningCustomer viewCollection operationsCampaign studioStaff audit trail
Commerce system

MrMaple storefront

I rebuilt MrMaple’s Shopify Empire theme around how people actually shop for plants. It has persistent mobile navigation, swipe-and-zoom product pages, USDA zone-aware care guidance, recently viewed plants, and product rails driven by Garden’s demand snapshots and feedback events.

The storefront stays simple and tactile; Garden handles the context behind it.

Shopify LiquidMobile PDPUSDA zonesCare guidanceDemand feedback
Visit the storefront →
Built and shipped

Ventful

Ventful was an iOS event app with discovery feeds, ticketing, and a recommendation engine that balanced personal taste, friends, and what was happening that night.

I co-founded it, designed the architecture, and shipped the product end to end to the App Store. Building it is what got me hooked on mobile.

iOSTicketingSocial feedsRecommendation engineApp Store