Brian – Machine learning, Data Science, Big Data

A multi-strategy hedge fund management firm that now focuses on delivering a financial platform for investors.

• Designed and built the end-to-end architecture for integrating multiple data sources into a centralized data warehouse from scratch, optimizing data flow, accessibility, and scalability to support business analytics and decision-making;
• Developed and deployed custom machine learning models and large language models (LLMs) from the ground up to classify and categorize report topics, significantly improving reporting accuracy and speeding up the delivery of actionable insights;

The world’s leading digital cross-device graph. It enables marketers to identify a brand customer or related household across multiple devices, unlocking critical use cases across programmatic targeting, media measurement, attribution, and personalization globally.

• Spearheaded a team of 5 engineers to design and implement SDK components, reducing model deployment time by 10% and increasing deployment frequency by 10%, empowering data scientists to experiment and productionalize models with greater efficiency.
• Engineered a high-throughput, request-response prediction architecture with P99 latency guarantees, scaling to handle up to 5,000 RPS.
• Implemented ONNX model support in Triton and Ray, slashing cold start times by 20% and significantly boosting inference responsiveness, enabling high-scale, low-latency predictions and ensuring seamless performance under heavy traffic.
• Integrated Datadog sidecar for real-time observability, enabling proactive monitoring of data drift and model degradation at scale, resulting in early detection and resolution of performance issues, ensuring continuous, high-quality model performance.
• Engineered a SDK to support zero-downtime updates and efficient routing for 1,000+ models, enabling rapid iteration to production.
• Mitigated cold start issues by preloading frequently accessed models, cutting latency by 20% during peak loads.
• Optimized model inference workflows by implementing advanced feature engineering and model optimization techniques, including pruning, reducing average prediction latency by 20% while preserving model accuracy.
• Mentored engineers on best practices for distributed systems and ML engineering, increasing team productivity.

Data & AI platform solutions for various IBM external clients across diverse industries for ensuring the scalability of their data and machine learning models.

• Spearheaded a team of 5 engineers to design and implement SDK components, reducing model deployment time by 10% and increasing deployment frequency by 10%, empowering data scientists to experiment and productionalize models with greater efficiency.
• Engineered a high-throughput, request-response prediction architecture with P99 latency guarantees, scaling to handle up to 5,000 RPS.
• Implemented ONNX model support in Triton and Ray, slashing cold start times by 20% and significantly boosting inference responsiveness, enabling high-scale, low-latency predictions and ensuring seamless performance under heavy traffic.
• Integrated Datadog sidecar for real-time observability, enabling proactive monitoring of data drift and model degradation at scale, resulting in early detection and resolution of performance issues, ensuring continuous, high-quality model performance.
• Engineered a SDK to support zero-downtime updates and efficient routing for 1,000+ models, enabling rapid iteration to production.
• Mitigated cold start issues by preloading frequently accessed models, cutting latency by 20% during peak loads.
• Optimized model inference workflows by implementing advanced feature engineering and model optimization techniques, including pruning, reducing average prediction latency by 20% while preserving model accuracy.
• Mentored engineers on best practices for distributed systems and ML engineering, increasing team productivity.

The team provided data modeling solutions to various external clients in multiple industries through IBM, addressing their specific business use cases.

• Collaborated with external business clients to align business objectives with tailored ML solutions, driving tangible outcomes. • Developed and implemented traditional machine learning models and neural networks, achieving a 15% average accuracy improvement and a 30% enhancement in model performance through effective feature engineering. • Created custom metrics like F1 scores and RMSE, leading to a 20% increase in stakeholder satisfaction and a 40% improvement. • Conducted A/B testing and statistical experiments, resulting in a 25% increase in user engagement for tested features. • Performed data cleaning on terabyte-sized datasets, enhancing predictive model accuracy and implementing retraining processes that improved performance by 10% over six months.

Canada Revenue Agency the revenue service of the Canadian federal government, and most provincial and territorial governments. The CRA collects taxes, administers tax law and policy, and delivers benefit programs and tax credits.

• Led and trained a baseline Machine Learning model to predict potential non-compliant tax payers across Canada recovering more than $1M.
• Gathered and established methodologies with auditors to ensure high-quality deliveries for different departments