Open to opportunities

Ariun Zaya

@ariunzaya

Message

I build AI systems, anomaly detection pipelines, and data-driven products.

Mongolia

Message

What I'm looking for

I'm looking to build practical AI and data products where I can apply machine learning, anomaly detection, business analysis, and product leadership to improve decisions and real-world outcomes.

I've built proprietary AI systems for Huchu Agro LLC that improve livestock tracking, predictive analytics, and herd-management operations. I also launched AI-driven e-commerce solutions with personalized recommendation engines to increase engagement and conversion.

At Spark Growth Inc, I architected an AI-driven anomaly detection and alerting pipeline for large-scale marketing data. I combined STL decomposition, Fourier signal processing, Triplet Autoencoders, and Extreme Value Theory to detect anomalies and support root-cause analysis.

I've applied machine learning to financial crime and public-interest data, including analyzing over 1 million transaction records for AML risk at the IMI BigData & AI Competition with Scotiabank. At the Centre for Community Energy Transformation, I used SQL, Excel, and Power BI to analyze household energy burden and create fundraising insights.

I'm completing an Honours BSc in Statistics at the University of Toronto, with minors in Computer Science and Business. My work has also earned Best in AI and overall second place at the DaVinci Hackathon for an AI-powered safety app for deaf users.

Experience

Work history, roles, and key accomplishments

Current

Google WTM Ambassador

Current

Google

Aug 2024 - Present (2 years 1 month)

Led community initiatives and events to promote STEM diversity, including hosting IWD Toronto 2025 for 200+ participants.

Event Planning Community Engagement Leadership

AI Software Developer

Spark Growth Inc

Sep 2025 - Jan 2026 (4 months)

Architected AI-driven anomaly detection pipeline for marketing data, using hybrid time-series models and EVT. Developed automated root cause analysis and conducted model evaluation.

Python Machine Learning Anomaly Detection Time Series STL Decomposition Fourier Analysis Deep Learning Triplet Autoencoders Extreme Value Theory Root Cause Analysis Model Evaluation

AI Engineer

Global Health Core Inc

Jan 2025 - May 2025 (4 months)

Enhanced Ollama AI model with emergency call button and refined prompt engineering. Assisted in model validation and performance testing.

Python Ollama Prompt Engineering LLM Model Validation

Machine Learning Developer

IMI BigData & AI Competition

Nov 2023 - Apr 2025 (1 year 5 months)

Developed ML models for AML anomaly detection on large financial datasets, improving accuracy by 20%. Designed hybrid clustering and generative models for customer representation.