Case Studies

The Problem

Watching a long video end-to-end to extract the useful part is slow. Most of it is skimmable, but sifting for the signal still costs the full runtime.

What I Built

A web product that turns a YouTube link into a clear, structured brief, with every summary kept in one workspace tied to a verified account. Google Gemini powers the summarization layer; Stripe subscription billing and credits were built in from the start rather than bolted on after.

Result

Full-length video becomes a skimmable brief in seconds, with a history that stays organized instead of scattered across notes.

The Problem

A Rightmove or Zoopla listing shows photos and an asking price. It doesn't tell a buy-to-let investor whether the investment actually works, so they cross-check a dozen scattered sources by hand, spending hours before deciding a property is even worth a viewing.

What I Built

A full-stack SaaS that takes a listing link and pulls 15+ verified UK data sources into one normalized view, then returns a single buy, hold, or avoid signal with yield calculations behind it. I built the data pipeline in PostgreSQL, the LLM analysis layer, Stripe billing, and the AWS deployment.

Result

Research that took an afternoon now takes seconds, and every underlying data point stays traceable, so the answer can be checked, not just trusted.

The Problem

A construction business was running projects across teams with completely different needs, including field crews, inventory, finance and management, with no shared system. Office and site had no single source of truth.

What I Built

An enterprise RBAC platform with six role-specific dashboards covering 15+ modules, plus a companion React Native app for teams working on site. I built the entire web frontend in React and MUI against a PostgreSQL backend on GCP.

Result

One system replacing fragmented, role-specific tracking, with real-time progress and financial oversight across every site.