Selected Work

Building AI products backed by strong machine learning foundations.

My work spans AI product development and applied machine learning. From building end-to-end AI applications to developing predictive models, I focus on turning technical capabilities into practical solutions that solve meaningful problems.

Featured Products

Currently in the works Takeon shift-continuity app for children's residential homes

🩺 AI Product / Healthcare

Takeon

Problem: In children’s residential homes, a support worker coming back from weeks of leave is often paired with agency cover who has never worked there. A verbal handover covers one shift, not three weeks, and the same incident gets written by hand three times at the end of a long shift.

I worked as a support worker in UK children’s homes and lived these gaps. Takeon gives anyone walking onto a shift a two-minute catch-up on each young person, covering exactly the days they were away. Staff talk through the shift once and Takeon drafts the handover, incident form and daily log in the young person’s own voice. A shift lead checks and publishes each record to one shared home record, and slow changes in eating, sleep and mood are flagged in plain language.

Product takeaway: I almost dropped this idea when I asked myself why staff couldn’t just hand over verbally. Questioning it surfaced the real gaps: continuity after leave, repeated paperwork, and patterns no single shift sees. In a setting this sensitive, the design choices that mattered most were about trust: AI only drafts, people approve, and the child’s words stay theirs.

AI Product Design LLM Drafting Voice Input Safeguarding by Design Claude Synthetic Data

Currently in development. A demo is coming soon.

Live product OpsSignal operational intelligence dashboard with anomaly detection, service health, and an AI investigation assistant Open product ↗

🚨 Featured AI Product

OpsSignal

Operational Intelligence Platform

17 Anomalies surfaced
92% System health
AI Investigation assistant
Overview Upload Data Anomalies Investigations Services Reports

Problem: Engineering teams often need to detect operational issues quickly and investigate incidents before customer impact grows.

What I built: An operator dashboard that turns telemetry into a morning briefing: upload data, detect anomalies, inspect affected services, and ask an AI investigation assistant for likely causes and next steps.

Product takeaway: The product stays useful only if the operator can move from signal to action. I designed around that loop: what is happening, which services are hit, and what to do next — with Isolation Forest detection and LLM-backed investigation still doing the underlying work.

TypeScript React Python FastAPI OpenAI Isolation Forest Docker Vercel

Product UI for the live prototype. Demo telemetry until a CSV analysis is completed.

Portfolio Pulse portfolio intelligence prototype

📈 AI Product / Product Builder

Portfolio Pulse

Problem: Investors often have portfolio updates, market signals, and company information spread across different places, making it difficult to quickly understand what needs their attention.

I designed and built Portfolio Pulse, an AI-powered portfolio intelligence prototype that helps investors bring company updates, market signals, and internal context into one place. The experience includes portfolio monitoring, company views, an attention queue, market signals, forecasts, and reporting.

Product takeaway: Building Portfolio Pulse helped me think about the full decision-making workflow: what information matters, how it should be prioritized, and how AI can help surface useful signals without overwhelming the user.

AI Product Design Cursor v0 Rapid Prototyping Vercel Product Analytics Synthetic Data

Demo environment using synthetic company and market data.

3D contact envelope illustration

Contact me

If you’re interested in working together, don’t be a stranger.

Say hi!

--:-- local time

Houston, Texas, USA