Richard
Marks

All learning is the result of thinking. This is a web application designed to address the problem of authenticating student thinking outside of the classroom, not just writing. Combining in-class writing baselines with student submitted digital work to create personalised, viva interviews powered by AI at scale. This is an ongoing project that is still in its validation phase.

A forecasting model for BonTon Bakery to predict sales and stock needs. The model uses historical sales data, weather data, school holidays and other factors to predict future sales and stock needs.

A CLI tool that uses AI to generate content for Github repositories based on the repositories metadata (README, commits and releases).

Starting as a Python script in 2023 (see below), I converted the project into a functional web application for wider use due to increased demand for the tool. See it here at https://batch.ac

Food ordering system for schools. Responsible for the frontend and backend development, focused on integrating payment with Stripe. Created the booking system for schools to order food for their students.

Batch is an AI-assisted reporting system I designed to solve a data consolidation problem in education. Teachers typically pull information from multiple fragmented sources and manually synthesise it into structured reports. I built Batch to automate that workflow: it ingests structured datasets alongside reference documents, applies LLM orchestration (via LangChain), and generates a full class set of consistent, tone-aligned reports. Turning a 6 hour manual process into a 1 hour automated one. Originally developed as a Python prototype in 2023.