Chamal Gomes

Summary

Principal data science leader with deep expertise in production ML systems, Python infrastructure, and LLM deployment. Proven record architecting end-to-end ML pipelines and enterprise-grade ML Ops in regulated healthcare. Delivers automation-first engineering and high-impact AI across clinical and financial domains.

💼 Experience

Healthscope - Data Science Lead

2022 Jan - Present

Principal Data Science Lead across 40+ hospitals: Architected and led the delivery of machine learning systems and analytics solutions within Australia’s second-largest private healthcare provider, enabling cross-team reuse, intelligent automation, and measurable business ROI.
Innovation for Healthcare Operations: Designed production ML systems that improved operational efficiency, patient care, and financial forecasting:
- Automated Staff Rostering 🔗: Developed a 10-week and 72-hour forecast model for automated staff scheduling across 160+ hospital wards.
- Automated Finance Forecast: Built episodic revenue model to support accurate finance forecasting and SOX-compliant journal posting.
- Revenue/Coding Optimization: Used LLMs to analyze EMR data and enhance patient care and revenue cycle performance.
- Preadmission clinic automation: Deployed NLP + LLM pipeline to identify high-risk patients for early intervention and improved care.
- Accounts Payable Automation: Used Azure Document Intelligence to streamline invoice processing and improve accrual accuracy.
ML Ops Platform Architect: Built AzureML infrastructure to enable secure and scalable ML experimentation and deployment. Including workspace isolation, role-based access, and compute orchestration—laying the foundation for long-term platform reuse and AI governance.
Advanced ML Engineering: Built mono-repo-based Python ML systems with SoTA tools, robust test frameworks and task automation. Emphasized clean architecture and infrastructure-as-code for production-grade deployment and maintainability.
Data Engineering and Dashboards: Designed DBT models feeding Azure Data Factory pipelines for deployed ML systems. Integrated PowerBI dashboards directly with model outputs to support decision-making across finance, clinical and ops.
Mentorship and Technical Uplift: Led internal workshops on test-driven scalable ML system design and experimentation patterns. Championed robust engineering practices and helped elevate junior team members’ Python and ML Ops proficiency.
Stakeholder Engagement: Collaborated with executive, finance, and clinical teams to align ML investments with strategic goals. Translated technical outcomes into business impact narratives to guide adoption and funding.

Healthscope - Data Analyst

2020 Jan - 2021 Dec

Systems Integration: Integration of core clinical and non-clinical hospital source application systems to establish a robust on-premises data warehouse, ensuring seamless data consolidation and accessibility. (WebPAS, Dimensions, Tech1, Riskman, etc)
Financial Modeling: Development of financial projection models utilised to identify market share by case-mix acuity and demography.
Advanced Predictive Analytics: Applied survival modeling techniques to analyze the probability distribution of hospital-acquired complications and patient readmissions, providing critical insights for risk management.
Machine Learning for Healthcare Operations: Designed revenue and clinical cost decomposition models accounting for inflation and case-mix changes, facilitating benchmarking of best practices across the hospital portfolio.
Operational Dashboards: Developed and maintained a comprehensive suite of operational healthcare dashboards, fostering a culture of evidence-based decision-making and driving operational improvements across hospitals.
Cloud Data Warehouse Migration: Designed and developed DBT models as part of a strategic cloud data warehouse migration project. Key models supported include patient invoicing, patient case-mix management, and admission-to-cash flow management.

NMG Consulting - Actuarial Intern/ Modeling Analyst

2018 Nov - 2019 Apr

Regulatory Reporting and Benchmarking: Managed Regulatory Risk-Based Capital (RBC) submissions for clients, contributing to the benchmarking of industry RBC data to ensure compliance and enhance strategic decision-making
Machine Learning in Fraud Detection: Collaborated with the Prediction Consulting team on a research initiative to apply machine learning methods for fraud detection, focusing on model design and performance evaluation
Comprehensive Reporting: Authored detailed reports documenting the functionality, advantages, and limitations of various machine learning models in the context of fraud detection, providing actionable insights for stakeholders.

💻 Key Skills

🛠️ ML Engineering

Tooling – Mono-repo infra (uv, ty, ruff)
QA – Pytest, coverage, static typing
Performance – Profiling, regression checks
Pipelines – Staged training, batch & streaming inference

🧩 System Design

LLM Workflows – RAG, fallbacks, prompt chaining
OSS Models – Integration, quantization, inference
Frameworks – LangChain, llama.cpp, Pydantic
Prototyping – PoCs with ReactJS & FastAPI

⚙️ MLOps & Infrastructure

Reproducibility – Nix, Docker
CI/CD – GitHub Actions, AzureML CLI
IaC – Terraform, AzureML workspace orchestration
Automation – just, component scaffolding

🧱 Data Platform & Analytics

Orchestration – Azure Data Factory (Snowflake + ML)
Feature Engineering – DBT-based SQL modeling
Dashboards – Power BI with integrated ML output
ETL Design – Composable, incremental patterns

🎓 Education

Bachelor of Commerce (Actuarial Science)

University of Melbourne, 2015 - 2017

Master of Commerce (Actuarial Science)(Research Pathway)

University of Melbourne, 2018 - 2019

Awards and Scholarships

University of Melbourne USA Travel Scholarship (San Francisco)
University of Melbourne Commerce Global Scholarship

📜 Research/Publications

Publication Statistics

Total Citations: 64 Google Scholar

Masters Thesis

2018 Jun - 2019 June

Insurance Fraud Detection with Unsupervised Deep Learning. Introduced a novel methodology for fraud detection and experience analysis using unsupervised deep learning models in collaboration with Centre for Actuarial Science at University of Melbourne and University of Hong-Kong.

Publications

2021

Gomes C, Jin Z, Yang H. Insurance fraud detection with unsupervised deep learning. Journal of Risk and Insurance. 2021 Sep;88(3):591-624, https://onlinelibrary.wiley.com/doi/10.1111/jori.12359

👥 Volunteer

University of Melbourne - Mentor

2023 Apr - Present

Providing mentorship for students within the faculty of business and economics with academic pathway guidance, career preparation, and sharing of experience to achieve their set goals and challenges.
🔗 View Program Information

Australian Catholic University - Tutor

2021 Jun - Present

Tutor at the ACU Clemente program for students with disadvantaged backgrounds.
Guiding students with weekly lectures, assignments, and reading material.
🔗 View Program Information