Hero

Hi!
I'm The Oxford Developer , a Backend (and aspiring Fullstack) Developer.

name:'TheOxfordDeveloper',
skills:['Python', 'Flask', 'AWS', 'HTML', 'CSS', 'JS', 'Numpy', 'OpenCV', 'Matlab'],
hardWorker:true,
quickLearner:true,
problemSolver:true,
hireable:function() {
return(
this.hardWorker&&
this.problemSolver&&
this.skills.length>=5
);
};
};

About me

Hi! I'm Chloe - an Oxford University Earth Sciences graduate with a Python-based integrated masters degree, as well as 2+ years of backend coding experience post-graduation.

I am particularly interested in supporting startups and have a wealth of experience in backend projects (including customer-facing support bots and generative AI chatbots, optical character recognition, and structuring unstructured data).

I am exploring frontend development and have recently built a chatbot user interface via Flask, JS, HTML, and CSS and deployed it as a fullstack web application (as well as this website, via React!).

TheOxfordDeveloper
Hero
Services
Hero

Technical Consulting - Backend Development

Example Work

Open to all backend development projects beyond examples listed below

  • Built an OCR pipeline to extract text from images, PDFs, etc
  • Developed tools to structure unstructured data into usable formats
  • Created a robust URL checker and web scraper for data validation
  • Integrated APIs and automated workflows to streamline operations
  • Designed and deployed scalable backend architectures

*Currently learning frontend development — this website is part of that journey!

Hero

Customer Support Chatbot (AI)

  • Built using trusted LLMs (Google's gemini-1.5-pro)
  • Fine-tuned on your company data
  • Understands natural language queries and context
  • Easy integration with websites
  • Custom UI and white-labelled for branded chatbot experiences
  • Support ticket email system to contact support team if the user would like more assistance
  • Continuously improves from real user interactions
Hero

Customer Support Chatbot (Non-AI)

  • Easy integration with websites
  • Custom UI and white-labelled for branded experiences
  • Lower operating costs for simple support needs
  • Uses mathematical formulas to match queries – 92% accuracy
  • Support ticket system for additional assistance
  • ❌ Does not learn from past interactions
  • ❌ Only retrieves pre-determined Q&A from database
Skills
PROJECTS

FAQ Support Bot

name:'FAQ Support Bot',
tools: ['Python', 'JS', 'HTML', 'CSS', 'Procfile', 'Flask', 'Heroku],
myRole:Full Stack Developer,
Description: Non-genAI retrieval-based support bot using Support Vector Machine and Cosine Similarity. Built with Python for the backend, Flask, HTML, CSS, JS and Procfile for the frontend, and deployed via Heroku. Designed to answer common queries customers have and reduce the amount of time the small business owner spends answering repetitive questions,
};

Generative AI Chatbot

name:'Generative AI Chatbot',
tools: ['Python', 'JS', 'HTML', 'CSS', 'Procfile', 'Flask', 'Heroku', 'Llama],
myRole:Full Stack Developer,
Description: Generative AI chatbot using Llama???. Built with Python for the backend, Flask, HTML, CSS, JS and Procfile for the frontend, and deployed via Heroku. Designed to simulate human-like conversations, offering dynamic, context-aware interactions that adapt to individual user needs, streamlining customer support, and improving engagement.,
};

Structuring Unstructured Data

name:'Structuring Unstructured Data',
tools: ['Python],
myRole:Backend Developer,
Description: Turned unstructured text data into a formatted, structured file of columns and rows using Regex (regular expression) patterns.,
};

Optical Character Recognition

name:'Optical Character Recognition',
tools: ['Python],
myRole:Backend Developer,
Description: Digitised physical books/text documents with Optical Character Recognition (OCR).,
};

URL checker/web scraper

name:'URL checker/web scraper',
tools: ['Python', 'Selenium],
myRole:Backend Developer,
Description: Designed to process a CSV file with around 30,000 URLs, checking whether each page contains relevant data by scanning for pre-specified messages. It operates in a headless browser environment using Selenium, with randomised user agents and session cookies to bypass bot protections. The output is a CSV file indicating whether a page contains useful data (i.e., the pre-specified message) (yes) or not (no). Key features include batch processing, captcha handling, and random delays to simulate human browsing.,
};
Hero
Background
Hero

2019 - 2023

Integrated Masters Degree - Earth Sciences

University of Oxford

Hero

2023 - present

Technical Consulting

Equine Match Ltd.

Hero

2024 - present

Independent Projects

GenAI & Non-GenAI chatbots, knowledge graph manipulation, building websites, building AWS applications

Blogs/Sample Project Code

1 months ago

0

1

Building and deploying a Non-Generative AI Support Chatbot with Flask

4 Min Read

Here are some images demoing the bot: Step 1: ask it a question Step 2: tell the bot whether...

1 months ago

1

Building a Headless URL Checker for Large-Scale Web Data Screening

3 Min Read

When dealing with massive datasets of URLs—especially in industries like sports analytics, finance,...

1 months ago

0

Parsing Unstructured Text Into Clean, Structured Data — A Regex-Powered Approach

2 Min Read

Transforming messy, human-readable text into clean, structured rows and columns is one of the most...

1 months ago

1

OCR: Digitise Physical Text Documents with Ease

2 Min Read

Easily convert scanned or photographed documents into editable, machine-readable text using Optical...

3 days ago

0

Building and deploying a Generative AI Support Chatbot with Flask and Google's Gemini API

5 Min Read

Example of what is being built: Introduction As a backend developer specialising in AI chatbots,...

Connect with me

If you have any queries, please reach out below!

chloe164@hotmail.co.uk

Oxford, United Kingdom

© Developer Portfolio by TheOxfordDeveloper