Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Hi, I’m John Toma, a passionate Software and Machine Learning Engineer based in Sydney. Over the years, I’ve had the privilege of leading exciting projects, including developing proprietary OCR and table extraction systems that significantly improve efficiency and reduce costs. I love tackling complex challenges and turning ideas into scalable software solutions. When I’m not coding or optimizing machine learning models, I enjoy managing teams and ensuring projects run smoothly from start to finish. I’m always keen to learn new technologies and collaborate on innovative projects that push the boundaries of what’s possible.…Hi, I’m John Toma, a passionate Software and Machine Learning Engineer based in Sydney. Over the years, I’ve had the privilege of leading exciting projects, including developing proprietary OCR and table extraction systems that significantly improve efficiency and reduce costs. I love tackling complex challenges and turning ideas into scalable software solutions. When I’m not coding or optimizing machine learning models, I enjoy managing teams and ensuring projects run smoothly from start to finish. I’m always keen to learn new technologies and collaborate on innovative projects that push the boundaries of what’s possible.

John Toma

Developer, Data Scientist, Full Stack Developer, +1





Hi, I’m John Toma, a passionate Software and Machine Learning Engineer based in Sydney. Over the years, I’ve had the privilege of leading exciting projects, including developing proprietary OCR and table extraction systems that significantly improve efficiency and reduce costs. I love tackling complex challenges and turning ideas into scalable software solutions. When I’m not coding or optimizing machine learning models, I enjoy managing teams and ensuring projects run smoothly from start to finish. I’m always keen to learn new technologies and collaborate on innovative projects that push the boundaries of what’s possible.…Hi, I’m John Toma, a passionate Software and Machine Learning Engineer based in Sydney. Over the years, I’ve had the privilege of leading exciting projects, including developing proprietary OCR and table extraction systems that significantly improve efficiency and reduce costs. I love tackling complex challenges and turning ideas into scalable software solutions. When I’m not coding or optimizing machine learning models, I enjoy managing teams and ensuring projects run smoothly from start to finish. I’m always keen to learn new technologies and collaborate on innovative projects that push the boundaries of what’s possible.

Available to hire

Hi, I’m John Toma, a passionate Software and Machine Learning Engineer based in Sydney. Over the years, I’ve had the privilege of leading exciting projects, including developing proprietary OCR and table extraction systems that significantly improve efficiency and reduce costs. I love tackling complex challenges and turning ideas into scalable software solutions.

When I’m not coding or optimizing machine learning models, I enjoy managing teams and ensuring projects run smoothly from start to finish. I’m always keen to learn new technologies and collaborate on innovative projects that push the boundaries of what’s possible.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Language

English

Fluent

Work Experience

Software & Machine Learning Engineer, Team Lead at Auxilis

March 1, 2022 - August 31, 2024

Led a research initiative to develop a proprietary OCR and table extraction system that achieved 90% of AWS Textract's accuracy on financial documents, cutting operational costs by 50%. Managed a team of ML engineers and up to 15 data labelers, overseeing project scheduling, budgeting, and execution. Designed and built a full-stack document processing system integrating Python ML engines with Node.js/Serverless backend, enabling flexible deployment options. Deployed Python-based ML engine on AWS ECS/Fargate using Docker, optimizing performance and reducing processing time by 80%. Developed advanced algorithms for spatial analysis, bounding box manipulation and text analysis improving OCR accuracy by 20%. Managed collection, synthetic data generation and curation of millions of images to enhance model robustness. Implemented 500+ unit tests achieving 98% test coverage, ensuring system reliability.

Systems Developer at Auxilis

November 1, 2021 - March 1, 2022

Developed and maintained full-stack bills management applications improving functionality and user experience. Created and refined UI components in Angular and Ionic framework, translating Figma designs into interactive user-friendly interfaces. Played key role in delivering major release of a cross-platform bill management app using Ionic Angular frontend and Serverless backend. Utilized AWS Amplify, CloudWatch and CloudPipeline to ensure smooth deployment and monitoring. Worked extensively with Angular, TypeScript, HTML/CSS and Ionic Framework to build scalable responsive applications.

Software & Machine Learning Engineer, Team Lead at Auxilis

March 1, 2022 - August 31, 2024

Led a research initiative to develop a proprietary OCR and table extraction system to replace AWS Textract, achieving 90% of Textract’s accuracy on financial documents and cutting operational costs by 50%. Managed a team of core ML engineers and data labelers, overseeing project scheduling, budgeting, and execution. Designed and built a full-stack document processing system with a Python-based ML engine and a Node.js/Serverless backend. Deployed the ML engine on AWS ECS/Fargate using Docker, optimizing performance and reducing processing time by 80%. Developed advanced algorithms for spatial analysis and text extraction that increased OCR model accuracy by 20%. Processed and curated millions of images, including synthetic data, enhancing model robustness. Fine-tuned models to improve inference speed and reduce misclassification. Implemented extensive unit testing to ensure system reliability.

Systems Developer at Auxilis

November 1, 2021 - March 1, 2022

Developed and maintained full-stack bills management applications, enhancing functionality and user experience. Created and refined UI components using Angular and Ionic, translating designs into scalable, responsive applications. Delivered a major release of a cross-platform bill management app using Ionic Angular and a Serverless stack. Utilized AWS services, including Amplify, CloudWatch, and CloudPipeline for smooth deployment and monitoring. Worked extensively with Angular, TypeScript, HTML/CSS, and the Ionic Framework to build efficient applications.