AI AUTOMATION · WORKFLOW DESIGN · INTERNAL TOOLS
Five years in ground-level operations taught me where the friction lives. I build systems to eliminate it.
I turn manual business processes (invoices, support queues, scheduling, internal knowledge) into working software and AI workflows. I start with the workflow rather than the technology, and I'll tell you when AI isn't the answer.
5 YEARS OPERATIONS: SALES, LOGISTICS, WAREHOUSE
REPEAT CLIENT: 2 ENGAGEMENTS
LIVE PRODUCTION BUILDS WITH REAL USERS
TRIPLETEN AI AUTOMATION CERTIFICATE · APRIL 2026
OPEN TO AI ROLES: REMOTE OR DFW
THE PROBLEM I SOLVE
Most businesses don't need AI. They need a process that takes 30 minutes instead of 3 hours.
- Invoice and document processing
- Customer support triage
- Sales operations and reporting
- Internal knowledge and search
- Scheduling and coordination
- Repetitive admin between disconnected systems
WHAT I BUILD
AI automation
Automating repetitive work involving documents, email, spreadsheets, forms, and business systems.
AI-powered internal tools
Focused applications that give a team a better way to manage information and make decisions.
AI knowledge systems
Connecting an organization's own documents to AI so people can find answers without digging.
AI-assisted customer operations
Support workflows where AI handles routine requests and escalates uncertain or sensitive ones to a person.
Workflow and process design
Mapping the process first, finding the bottleneck, then deciding whether automation or AI actually helps.
SELECTED WORK
Client operations automation
BUSINESS PROBLEM
A repeat client was handling sales operations, data analysis, and prospecting by hand across spreadsheets and disconnected tools.
WHAT I BUILT
Two separate engagements for the same client: a field sales tool and a data analysis workflow. Each was scoped to one specific bottleneck rather than a platform rebuild, and each was maintained after delivery.
RESULT
Two engagements with the same client, including a tool in daily use for several months and follow-on work currently being scoped.
- STATUS
- CLIENT WORK: LIVE
- TIMELINE
- FEB 2026 – PRESENT
- ROLE
- Problem definition · Workflow design · Application development · Client communication · Maintenance
- ARCHITECTURE
- Two independent applications, each scoped to a single workflow, integrating with the client's existing spreadsheets rather than replacing them.
- STACK
- React · Supabase · Lovable · Spreadsheet integrations
Friendly Friday Hike (fridayhikes.com)
BUSINESS PROBLEM
A women's hiking group was coordinating monthly hikes, RSVPs, and attendance by hand across text threads and email.
WHAT I BUILT
A member-facing site with passwordless sign-in, a published season schedule, self-service RSVP that members can change, self-marked attendance history, and automated reminder emails. Admins control roster visibility from a settings toggle rather than needing a code change.
RESULT
Live with 30 registered members and the full season published for RSVP.
- STATUS
- CLIENT WORK: LIVE
- TIMELINE
- AUG 2026 – PRESENT
- SCALE
- 30 members · 8 scheduled events
- ROLE
- Problem definition · Workflow design · Application development · Deployment · Client communication · Maintenance
- ARCHITECTURE
- Passwordless authentication over a managed backend, with scheduled reminder emails and admin controls driven by a settings toggle rather than a code change.
- STACK
- Lovable · Lovable Cloud (auth, database, email)
Corkboard (corkboard.wine)
BUSINESS PROBLEM
A private network of wine enthusiasts wanted a place to rate wines already tried and track the ones still on the list, something more personal than a general-purpose rating app.
WHAT I BUILT
A wine-logging app, invite-only by design, with an uncapped invite chain doing the rest of the growth: a small seed group gets a code, and from there the network grows through invites those users choose to send, with no cap set from the top.
RESULT
Live and invite-only. Three seed invite codes went out at launch, and each holder is free to invite as many people as they want, so the initial cohort is a seed for however far the network wants to grow on its own.
- STATUS
- CLIENT WORK: LIVE
- TIMELINE
- AUG 2026 – SEP 2026
- ROLE
- Problem definition · Product design · Application development · Deployment · Client communication · Maintenance
- ARCHITECTURE
- Invite-only entry flow logs wines through a progressive drill-down search (brand → varietal → vintage) into a Letterboxd-style poster-grid feed, backed by a client-owned Supabase database with Resend-delivered auth and invite email rather than a platform-managed backend.
- STACK
- Claude Code · Supabase · Resend · Cloudflare
Invoice processing automation
BUSINESS PROBLEM
Finance teams manually retype information from incoming invoices into spreadsheets and accounting systems.
WHAT I BUILT
An automated pipeline that reads invoice PDFs, extracts the key fields, checks whether the extraction is confident enough, routes uncertain invoices to a person for review, and writes approved records into a spreadsheet.
RESULT
Demonstrates a production-shaped document workflow with confidence gating and human review. Built and tested against sample invoice PDFs, not live production invoices.
- STATUS
- PROTOTYPE
- TIMELINE
- FEB 2026
- ROLE
- Workflow design · Architecture · Prompt engineering · Evaluation & testing
- ARCHITECTURE
- A document extraction pipeline with a confidence gate that splits output into an automatic write path and a human review queue.
- STACK
- UiPath · Document Understanding · Claude · Google Sheets
ALMAGEST: personal knowledge system
BUSINESS PROBLEM
Notes, sources, and research accumulate faster than anyone can organize them, and asking an AI about your own material usually means it can't tell a solid source from a weak one.
WHAT I BUILT
A knowledge system with a conversational retrieval interface, a 3D map of how the knowledge connects, and a trust-scoring layer that rates how well each retrieved source can actually support the claim it's being used for.
RESULT
In daily personal use. Development surfaced and fixed a real defect: a source-deduplication bug that let one document be counted several times and inflate a claim's confidence score. The first assertion-based test suite in the project caught it.
- STATUS
- LIVE: PERSONAL BUILD
- TIMELINE
- JUN 2026 – PRESENT
- SCALE
- 202 sources · 5 backend services
- ROLE
- Architecture · Prompt engineering · Application development · Evaluation & testing · Deployment
- ARCHITECTURE
- Retrieval-augmented generation over a vector store, with a separate trust-scoring layer that rates each source's support for a claim independently of the answer.
- STACK
- React · Supabase (pgvector, edge functions) · Claude API · OpenAI embeddings
AI customer support workflow
BUSINESS PROBLEM
Support teams spend most of their time answering the same routine questions, while genuinely difficult cases wait in the same queue.
WHAT I BUILT
A support workflow design where AI answers routine requests from a knowledge base, checks its own confidence before responding, escalates uncertain or sensitive cases to a person, confirms satisfaction, and raises priority when a customer isn't satisfied.
RESULT
A worked design for human-in-the-loop support: routing rules, confidence thresholds, escalation of uncertain or sensitive cases, and satisfaction follow-up. This is a designed workflow, not a deployed product.
- STATUS
- PROTOTYPE
- TIMELINE
- JAN 2026
- ROLE
- Problem definition · Workflow design · Architecture
- ARCHITECTURE
- Confidence-thresholded routing between automated response and human escalation, with a satisfaction check that raises priority on a negative result.
- STACK
- Zapier · n8n · VAPI · Glean
BracketGenius
BUSINESS PROBLEM
A test of whether careful prompt design and published analytics frameworks can beat naive prediction on a hard forecasting problem.
WHAT I BUILT
An NCAA tournament prediction system built on published basketball analytics frameworks, run live through the full 2026 tournament with bracket tracking.
RESULT
73% game accuracy across 63 games, against a roughly 75% benchmark from established models. One tournament is a small sample and the approach carries real overfitting risk. The case study explains both rather than treating the number as settled.
- STATUS
- LIVE: PERSONAL BUILD
- TIMELINE
- FEB – MAR 2026
- SCALE
- 63 games · 73% accuracy
- ROLE
- Problem definition · Prompt engineering · Evaluation & testing · Deployment
- ARCHITECTURE
- Published analytics frameworks applied through structured prompts, evaluated game by game against live tournament results.
- STACK
- Lovable · Claude API
HOW I WORK
- 01Understand the workflow
- 02Find the bottleneck
- 03Decide whether automation or AI actually helps
- 04Build a focused solution
- 05Test the failure cases
- 06Deploy and improve
Step three is the one most AI projects skip. Sometimes the answer is a spreadsheet formula and a changed handoff, and I'll say so.
AI AND OPERATIONS
AI skills are only useful if you understand the work.
There are a lot of people who can assemble an AI demo. Far fewer have worked inside the environments they're trying to automate. I spent five years in field sales, logistics, and warehouse operations before I built any of this.
- SALES
- customer workflows, account management, field productivity
- LOGISTICS
- routing, scheduling, handoffs, operational data
- WAREHOUSE OPERATIONS
- repetitive processes, throughput, inventory, coordination
I don't start with "where can we put AI?" I start with "where is the business losing time, money, or attention?"
FOR RECRUITERS
Looking for someone who can bridge AI and operations?
- AI AUTOMATION
- AI WORKFLOW DESIGN
- INTERNAL AI TOOLS
I build AI workflows and applications, and I came to it from five years of operational work in sales, logistics, and warehouse environments. I'm based in Arlington, Texas and open to full-time, contract, or temp-to-hire roles, remote or in the Dallas-Fort Worth area, with relocation possible for the right full-time role.
FOR BUSINESS OWNERS
Have a process that takes too much time?
You don't need to know whether you need AI. Tell me what your team currently does by hand, and I'll tell you whether automation, AI, a small application, or nothing at all, is the right answer.