I turn messy workflows into AI systems teams actually use.
I find where revenue gets delayed, operating costs accumulate, and teams hit capacity. Then I build practical automations, agents, and internal tools that improve the economics of how work gets done.
Featured Project
Semantic Visual Search for Creative Teams
Replaced chaotic Dropbox searches with a semantic, vibe-based search experience for 100k+ design assets.
Stack: Next.js/React · TypeScript · FastAPI · Vertex AI · Pinecone · AWS
🧠 What I Learned
Impact of AI-powered search infrastructure on team productivity: embedding models, vector databases, and scalable systems that eliminate asset discovery friction
✅ Outcome
Validated → scaled → became daily workflow
What I Build
Work Projects
AI-Native Email System
Building a job-based AI system for Sock Club's sales inbox. It turns conversations into shared context, identifies the work inside them, and advances each job only as evidence supports more autonomy. The first workflow is live; the universal router runs in shadow mode.
Quote Agent
An AI agent that prepares custom sock and packaging quotes from an @mention in Front. It retrieves live customer and product data, applies account-specific pricing rules, and generates a PDF for the sales rep to review, reducing manual work and time spent waiting in the sales support queue.
Portal+ Assist
An AI-assisted Front plugin connected to our internal portal. Sales teammates can submit customer design requests without leaving their inbox, with AI turning emails and attachments into clear design briefs that help reduce rework from missed context.
Personal Tools & Experiments
🔄 Reflow
Discover how work actually happens through privacy-first browser observation and AI-powered task inference.
📰 Personal Content Engine
An AI-assisted workflow that turns newsletter research into source-backed content briefs, with staged LinkedIn drafting and an editorial feedback loop.
✨ AI Button Share Generator
Generate embeddable AI share buttons for ChatGPT, Perplexity, and Google AI with custom prompts.
About Me
I bridge the gap between operations and engineering.
Every company has craft. AI only compounds it when the workflow is clear. The future belongs to teams that can turn expertise, handoffs, and judgment into coordinated systems people trust enough to use.
My current work is internally forward-deployed: I sit close to teams, study how work actually moves, spot the friction they've learned to ignore, and ship AI-enabled tools that fit real constraints.
By title, I'm a Sales Operations Manager, but I've been building tech for over a decade. I graduated from the University of Texas at Austin with a degree in Economics & Business, completed UT's full-stack web development bootcamp, and spent 13 years in startups across multiple industries, shipping products like a top-grossing #13 U.S. Health & Fitness app.
Today, I'm focused on workflow systems that multiply people: internal search, triage assistants, intake flows, and micro-apps that remove bottlenecks while keeping humans accountable for the calls that matter.
I use AI pair-programming, rapid prototyping, and modern cloud tooling to turn workflow problems into working systems fast. Whether I'm using Python or TypeScript, building in AWS or GCP, I rely on agents, observability, and tight feedback loops over theory.
By prioritizing revealed behavior over stated needs, I ensure I'm building for the 20% of signals that drive 80% of the results. This prevents "feature bloat" and ensures the tools I build actually stick.
My goal: build operating leverage. Systems that protect revenue, reduce avoidable cost, and help teams handle more without adding unnecessary complexity or headcount.
The Last Mile of AI
“Enterprises buying AI are like your grandma getting an iPhone. They want to use it. They just need someone to set it up.”
The setup is the work.
I map how work actually happens, connect the tools and data, build the AI layer, and stay close until the system earns enough trust to become part of the workflow.
Approach / Philosophy
How I think about building internal tools
| Step | What I Do | Mindset |
|---|---|---|
|
1️⃣
Discover
|
Shadow workflows; find the gap between what people say and do. | Behavior > Talk |
|
2️⃣
Prototype
|
Build a 48-hour "vibe check" using real data. | State-it-back test |
|
3️⃣
Validate
|
Track adoption; ensure humans stay in the loop for one-way doors. | Judgment > Automation |
|
4️⃣
Scale or Kill
|
Double-down or move on; never automate a broken process. | Leverage or nothing |
Builder Stack
The stack I use to prototype, ship, and operate AI-assisted internal systems for real teams.
🧠 AI Development Core
AI-native IDE for fast implementation, deep refactors, and context-aware debugging across full project codebases.
Codex / Claude Code
Coding Agents
Coding agents that help build features, fix bugs, and refactor code with context from the whole project.
Deployment and edge platform for shipping Next.js apps quickly, with built-in support for AI SDK workflows and scalable delivery.
Cloud IDE for fast full-stack iteration and clickable prototypes, de-risking direction before production engineering.
Background job orchestration for long-running AI tasks, retries, schedules, and event-driven workflows in a TypeScript stack.
"The best AI tools are the ones that fade into the background. You focus on building, they handle the complexity."
Writing
Stories and frameworks from building tools, services, and products
🧩 Building an AI-Native Email System, One Job at a Time
How I'm building an AI-native email system at Sock Club through job-based workflows, shadow mode, and human-reviewed evaluations.
🌅 Building Semantic Search for 100k+ Creative Assets
How I transformed a chaotic Dropbox archive into an AI powered search system that understands concepts instead of filenames.
🧠 Why Every Company Needs an Internal Tools Builder
How high-agency operators, "vibe coding," and AI are redefining how work scales inside organizations. Strategic insights on the future of operational leverage.
⚙️ How to Productize HubSpot Services
How we turned one-off projects into scalable, recurring value. Learn the framework, challenges, and results from building Trujay's HubSpot Admin Program.
🎯 Designing Customer Experience Systems
A practical framework with real results from the HubSpot Admin Program. Learn how to map touchpoints, standardize delivery, automate processes, and measure outcomes.
🏋️♂️ Building Fitness Apps: Lessons From My First Startup
The challenges faced and overcome while building my first startup. From Flexter to Toned (#13 top-grossing Health & Fitness app), learn how we went from zero experience to 300k+ downloads and $300k+ in sales.
Workflow Friction Audit Template
Find the workflows worth fixing first before you invest in automation, agents, or internal tools.
Use it to score workflows by: