Xan Merritt
Verified Expert in Engineering
ML Engineer and Developer
Halifax, NS, Canada
Toptal member since September 25, 2026
Xan is an independent ML engineer and consultant with 12+ years of experience building out AI across social media, fashion, and event ticketing. Xan specializes in RecSys and NLP/LLMs. While under contract with HyperComply, they delivered over 15% model accuracy and over 55% audit efficiency. Xan is production-focused, pragmatic, and ethically-centered.
Portfolio
Experience
- Software Engineering - 15 years
- Data Science - 10 years
- Python - 10 years
- Scala - 10 years
- Natural Language Processing (NLP) - 10 years
- Artificial Intelligence (AI) - 10 years
- Machine Learning - 10 years
- Open Neural Network Exchange (ONNX) - 7 years
Preferred Environment
Docker, Python, Machine Learning, Data Science, Scala, Open Neural Network Exchange (ONNX)
The most amazing...
...recommender system I've built improved user engagement for a ticket sales platform by leveraging language models and solving the cold-start problem.
Work Experience
ML Engineer and Data Scientist (Consultant)
Self-employed
- Worked as an independent machine learning engineer and data scientist consultant on multiple projects.
- Developed an end-to-end recommender system for a ticket sales platform. Performed custom development, analysis, and tuning to support production deployment of the universal recommender system built on Apache projects.
- Developed an end-to-end reactive streaming online machine learning pipeline for content-based Twitter follower and retweet count regression using Akka Streams, Apache Kafka, Apache CouchDB, Apache Spark Streaming/ML, and PMML.
Member and ML Engineer (Consultant)
Tribe AI
- Worked as a member and machine learning engineer consultant on various projects remotely through Tribe AI.
- Built an automated security questionnaire auditor (for HyperComply under contract, via Tribe AI).
- Implemented the Tribe solution as a custom recommender service. Offline testing showed results that were notably better than those of the black-box AWS alternative.
ML Engineer and Data Scientist (Consultant)
ActionML
- Developed a page variant recommender service based on contextual bandits using Vowpal Wabbit and Apache PredictionIO for use in an eCommerce platform.
- Performed custom development, analysis, and tuning to support production deployment of an end-to-end recommender system for streaming video using Apache PredictionIO and Apache Mahout.
- founded ActionML along with other ex-PredictionIO folks.
Senior Data Scientist
Salesforce Marketing Cloud
- Developed a technical stack for real-time analytics infused with machine learning and NLP using distributed storage, in-memory compute, NewSQL and graph databases, stream processing, and multi-GPU systems.
- Worked on applications in sentiment analysis, relevance determination, topic modeling, insights extraction, trending detection, and demographics for social media listening, publishing, and engagement.
- Led the first data science team within Salesforce Marketing Cloud.
Java Developer, MTS, Software Engineering
Salesforce Marketing Cloud
- Oversaw a self-organizing team of 4 developers. Shepherded a large-scale integration project from conception through technical preview to market: a scalable, fault-tolerant message routing, data enrichment, and delivery server.
- Acted as a founding team member and senior developer on a scalable, fault-tolerant integration project handling large volumes of social data.
- Worked as a mid-tier developer focusing on Java DAO layer work for the insights data integration project and language classification at Radian6 until the Salesforce acquisition.
Experience
HyperComply
The increase in accuracy significantly improved the audit team's throughput, with team efficiency rising by 55%. In addition to the model improvements, Tribe AI also built the foundation of an ML infrastructure that would allow the team to ingest new data, retrain models rapidly, and deploy at scale.
Education
Master's Degree (ABD) in Artificial Intelligence
University of Amsterdam - Amsterdam, Netherlands
Bachelor's Degree in Cognitive Systems
The University of British Columbia - Vancouver, Canada
Skills
Libraries/APIs
Pandas, TensorFlow, PyTorch
Tools
Open Neural Network Exchange (ONNX)
Languages
Python, Java, Scala, C++, JavaScript, C
Frameworks
Apache Spark
Platforms
Docker, AWS IoT
Other
Machine Learning, Data Science, Natural Language Processing (NLP), Machine Learning Infrastructure, Software Engineering, Artificial Intelligence (AI), Model Evaluation, Model Deployment, Statistical Modeling, Recommendation Systems, Model Tuning, Quantization, Personalization, Model Monitoring, Machine Learning Operations (MLOps)
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring