Engineering.
Operations.
AI Builder.

I build AI tools, agents, automation workflows and production ready apps. Engineering background, a decade running operations — which means I understand how organisations actually work before trying to automate them.
Currently building and shipping products in public — learning what it actually takes to go from zero to paying users.

"I spent a decade running hotel operations before I wrote my first line of code. That combination turns out to be useful."

My career started in hospitality — front of house, operations, the kind of work where you solve problems in real time with the people right in front of you. A decade in, I taught myself to code. Not because I wanted to become a developer, but because I have always been restlessly curious about how things are built. I never stopped.

Today I am an Engineering People Manager at a fully remote fintech company, leading eleven engineers across three teams. I also build AI-assisted internal tooling, lead the company-wide SDLC rollout, and design delivery frameworks. It is the job where seven years of operational leadership and a software engineering background finally make sense together.

I am completing a Master's in Generative AI Engineering and building production ready products in public. I work across Claude, Gemini, GPT, DeepSeek, Kimi, GLM, and others via OpenRouter — learning the best use cases for each model and building my own custom multi-model workflows. I switch often and adapt fast.

What I'm shipping

Three projects. Built solo, AI-assisted end to end.

Find the roles that actually fit. Automatically.

You build your profile once. The agent runs daily — scraping job boards, scoring each listing against your profile using AI, and sending a formatted digest of the best matches.

It learns from what you save and discard. Every action feeds a signal into Redis. Over time the agent builds a preference model from your behaviour and adjusts future scoring accordingly.

Also generates a clean ATS-proof CV from your profile in multiple styles. Build your profile once, export anywhere.

I built it because the problem was real and nobody had solved it the way I wanted.

In production · active development

Next.js · TypeScript · Neon PostgreSQL · Drizzle ORM · Upstash Redis · Gemini · Resend · Vercel

Zero to ten paying customers

I kept reading GTM advice that assumed you already knew what you were doing. I did not. So I built a framework — seven stages from idea validation to first ten paying customers — and then built the app around it.

GTM OS walks you through each stage, forces the right questions, and keeps you honest about where you actually are versus where you think you are.

Now it is getting a team — seven AI advisors and four simulation agents that read your project context and remember what you decided. A framework tells you what stage you are in. A team tells you if your thinking is solid.

In use · being tested on CV Tailor

Next.js · Supabase · Gemini API

Training plans built around your body, not a generic athlete.

FitnessHub treats the menstrual cycle as a first-class training input, not a wellness side panel. Every session is calibrated to your cycle phase — notes on effort, joint care, and load. Built around how a specific body works, not a generic template with tracking bolted on.

Currently live as a single-user app built around a 30-week half marathon plan. Training calendar, HR zone targets, strength circuits, cycle tracker with per-phase guidance, and body metrics.

Next phase: AI-generated plans across any distance and timeframe, multi-user support.

Live · Single user · Returning soon

Next.js · Supabase · Tailwind · Recharts · dnd-kit · Vercel

Get in touch

If you want to talk about AI tools, automation, business operations, or what I'm building, use the form.

By clicking Send message you accept the aviso legal and privacy policy.