Jason Cherry
Verified Expert in Engineering
Machine Learning Engineer and Developer
Spokane, WA, United States
Toptal member since March 6, 2026
Jason is a senior data scientist and ML engineer with 10+ years of experience, with 6+ years in production systems. He specializes in forecasting, anomaly detection, causal inference, and LLM-driven decision systems. Bringing a strong Bayesian and econometrics foundation with depth in inference under uncertainty, Jason rapidly embeds with client teams and translates ambiguous requirements into deployable solutions. He builds models that reach production and reduce cognitive load.
Portfolio
Experience
- Python - 12 years
- Data Science - 10 years
- Statistical Modeling - 10 years
- Predictive Modeling - 9 years
- Machine Learning - 9 years
- Time Series Forecasting - 6 years
- Azure - 5 years
- Machine Learning Operations (MLOps) - 4 years
Preferred Environment
Python, PyTorch, Scikit-learn, Pandas, SQL, FastAPI, Docker, Kubernetes, Amazon Web Services (AWS), PyMC
The most amazing...
...system I've built applied forecasting and anomaly detection to energy telemetry across hundreds of buildings, cutting manual monitoring from hours to minutes.
Work Experience
Senior Data Scientist
Pacific Northwest National Laboratory
- Developed anomaly detection and graph analytics models that surfaced previously unobservable behavioral patterns, improving analyst triage efficiency across large-scale distributed infrastructure.
- Built a production LLM-driven remediation system that matched incoming vulnerability data against system context, generated ranked recommendations with calibrated confidence scores, and routed outputs between automated approval and human review.
- Designed and implemented production ML pipelines for large-scale telemetry analysis, enabling continuous automated monitoring and anomaly detection at scale.
Senior Data Scientist
Aquicore
- Designed and implemented the organization's first MLOps platform, growing production models from near zero to a dozen in 12 months and enabling independent model development, automated testing, and continuous deployment.
- Developed forecasting models on large-scale building energy telemetry streams, enabling proactive detection of abnormal energy consumption patterns across hundreds of commercial buildings.
- Operationalized machine learning (ML)-driven building monitoring systems that transformed raw energy telemetry into automated alerts and actionable operational insights.
- Developed a human-in-the-loop workflow integrating ML forecasts and anomaly detection with operator feedback to improve model accuracy and operational usability.
Data Scientist
RiskLens
- Co-developed a patented (US 12019755) statistical methodology for quantifying cyber risk under sparse data conditions, validated through Monte Carlo and Markov Chain Monte Carlo simulation.
- Built Bayesian risk models applying probabilistic updating to support decision-making under uncertainty, producing calibrated outputs that communicated both point estimates and confidence intervals to non-technical stakeholders.
- Applied graph-based process mining to large-scale clickstream data to uncover behavioral patterns and workflow bottlenecks, informing the design of an in-product tutorial system.
Data Engineer - Contract
Cambium Defense Systems
- Built HIPAA-compliant ETL pipelines on PostgreSQL, processing hundreds of patient records daily, delivering the client's first automated clinical risk scoring workflow with standardized audit logging.
- Engineered time-decay feature transformations to correct temporal bias in clinical risk models, improving prediction stability across rolling cohorts.
- Translated clinical risk scoring requirements from non-technical stakeholders into pipeline architecture and schema design, scoping the workflow end-to-end against a fixed contract deliverable.
Data Scientist
University of Colorado
- Developed a recommender system to prioritize alumni engagement opportunities, achieving approximately 95% alignment with expert evaluation.
- Built annual and monthly forecasting models for fundraising performance, enabling data-driven planning and campaign management.
- Developed churn prediction models to identify at-risk alumni and support targeted retention and engagement strategies.
- Presented technical research at PyData Denver in 2017 and 2018, delivering talks on regularization methods and applied TensorFlow to a practitioner audience.
Program Evaluator
Jewish Family Service of Colorado
- Designed statistical analyses evaluating outcomes across multiple social service programs, supporting funder-mandated evaluations and continued program funding.
- Developed data reporting systems enabling program leadership to monitor performance and outcomes across multiple service initiatives.
- Led migration from legacy databases to modern data systems supporting approximately 150 staff, completing the transition with zero operational downtime.
Experience
PertDist
https://github.com/Calvinxc1/PertDistmv-laplace
https://github.com/Calvinxc1/mv-laplaceNew Eden Analytics
https://github.com/New-Eden-Analytics/OverviewVector-Graph Agent Memory
https://github.com/Calvinxc1/vector-graph-memoryArgo Spatiotemporal Interpolation
https://github.com/Calvinxc1/argo-data-interpolationEducation
Bachelor's Degree in Computer Science
Western Governors University - Salt Lake City, UT, USA
Master's Degree in Public Policy
Brandeis University - Waltham, MA, USA
Skills
Libraries/APIs
PyTorch, Scikit-learn, Pandas, SciPy, API Development, Claude API, OpenAI API, PyMC
Tools
Terraform, Amazon SageMaker, Claude Code
Languages
Python, SQL, Snowflake
Paradigms
ETL, Anomaly Detection, Automation
Storage
Databases, Data Pipelines, Database Management, JanusGraph, PostgreSQL
Platforms
Docker, Kubernetes, Amazon Web Services (AWS), Azure, Ollama
Frameworks
Agentic Frameworks
Other
Statistical Modeling, Machine Learning, Data Analytics, Data Science, Predictive Modeling, Data Analysis, Statistical Analysis, Statistics, FastAPI, Econometrics, Algorithms, Software Engineering, Time Series Analysis, Time Series Forecasting, Forecasting, Graph Analytics, Probability Theory, Numerical Modeling, Data Engineering, API Integration, K3s, Open Source, Data Classification, Logistic Regression, Linear Regression, APIs, Time Series, Integration, Software Architecture, Third-party Integration, Architecture, RESTFul APIs, CI/CD Pipelines, Multivariate Statistical Modeling, Cloud Platforms, Data Modeling, ETL Development, Model Evaluation, Causal Inference, Distributed Systems, Bayesian Statistics, Markov Chain Monte Carlo (MCMC) Algorithms, Recommendation Systems, Program Evaluation, Data Visualization, Reporting, Scientific Computing, Machine Learning Operations (MLOps), AI Architecture, Artificial Intelligence (AI), Retrieval-augmented Generation (RAG), Large Language Models (LLMs), Prompt Engineering, Agentic AI, Agentic RAG Systems, LangChain, Vector Databases, AI Agents, AI Automation, AI Design, AI Programming, NixOS, Qdrant, Probabilistic Modeling, PydanticAI, Quantitative Modeling, Risk Management, RAG Systems, RAG Pipelines, AI Integration, Anthropic, Opus, Sonnet, OpenAI, Generative Pre-trained Transformer 4 (GPT-4), OpenAI GPT-4 API, Full-stack, AI Pipeline, Natural Language Processing (NLP), Agentic AI Systems
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