Open to opportunities

Christabelle Pabalan

@christabellepabalan

Message

Healthcare AI Lead Data Scientist specializing in hybrid ML/LLMs, evaluation, and Responsible AI for clinical impact.

Zimbabwe

Message

What I'm looking for

I’m looking to lead healthcare-focused AI teams building production-ready hybrid ML/LLM systems, with rigorous AI evaluation and Responsible AI practices, to deliver measurable improvements in clinical operations and patient outcomes.

I’m a Lead Data Scientist specializing in healthcare AI, machine learning, and generative AI, with experience architecting production AI platforms that improve clinical operations and outcomes. At Qventus, I led AI strategy for perioperative care coordination, building hybrid machine learning and generative AI for cancellation prediction (accuracy 0.67→0.79; specificity 0.71→0.91) and a hybrid ASA classification model that improved performance by 76% over XGBoost while operating within 97% of the interrater agreement ceiling.

I also developed an appointment duration model that reduced screening waste by 83% and saved 10.8 staff hours per 100 patients. My work extends into LLM evaluation and continuous improvement—introducing autonomous prompt-optimization agents, golden datasets, LLM-as-judge evaluation, and human-in-the-loop validation—paired with Responsible AI through fairness-aware dataset design and evaluation standards. I provide technical leadership by mentoring data scientists and partnering with product, engineering, and stakeholders to define AI strategy.

Experience

Work history, roles, and key accomplishments

Current

Lead Data Scientist

Current

Qventus

Sep 2023 - Present (2 years 10 months)

Led AI strategy for Qventus's perioperative care coordination platform, architecting production hybrid ML/LLM systems to improve surgical planning and operational efficiency. Built models for cancellation prediction and appointment duration, and developed an LLM evaluation platform with autonomous prompt-optimization agents and human-in-the-loop Responsible AI practices.

Hybrid Machine Learning Generative AI Cancellation Prediction predictive modeling Appointment Duration Modeling LLM Evaluation Prompt Engineering Golden Datasets Human In The Loop Responsible AI Fairness Aware Dataset Design Clinical AI Evaluation

Machine Learning Engineer II

Headspace Health

Nov 2021 - Jul 2023 (1 year 8 months)

Developed machine learning solutions for Headspace Health, including a recommendation system wrapper to improve content discoverability and engagement. Optimized clinician intake predictions using a churn model to increase delivered appointments and reduce unbooked rates.

Recommendation Systems predictive modeling Churn Modeling Clinician Intake Prediction

Data Scientist

Loop

Jul 2021 - Nov 2021 (4 months)

Collaborated with a clinical team to identify customer interactions that maximize skill conversion likelihood and mental health metrics. Deployed a random forest model to predict the most likely skill sign-up and designed/streamlined A/B testing analysis.

Random Forest predictive modeling A B Testing

Machine Learning Researcher

UCSF Radiology

Jan 2021 - Sep 2021 (8 months)

Built a multi-task U-Net for Alzheimer’s-related MRI analysis, combining segmentation with regression to predict ADAS-cog scores. Improved segmentation accuracy and increased the regression r2 score by integrating demographic data using a novel approach.

U Net MRI Regression Modeling Alzheimer’S Outcomes Prediction PyTorch Pytorch Lightning Nibabel scikit learn