Applied Data Scientist
GAINSystems
Applied Data Scientist
Applied Research Group – Supply Chain Optimization
About GAINS
GAINS is on a mission to make supply chains smarter, faster, and self-improving, powered by AI. Our decision intelligence platform doesn't just support decisions, it drives them by aligning strategy, planning, and execution across every level of the supply chain. We serve inventory-intensive industries where the stakes are high and the complexity is real, helping customers move from reactive, spreadsheet-driven planning to continuously learning, AI-led operations that deliver measurable results fast. At GAINS, we call it Moving Forward Faster— and it's not a tagline, it's how we're redefining what's possible in driving supply chain decisions.
About the Role
As an Applied Data Scientist on the Applied Research Group at GAINS, you will research, design, build, and deploy production ML models that directly improve supply chain outcomes for enterprise customers. This is a hybrid role that spans the full ML lifecycle—from exploratory analysis and model development through production deployment and ongoing performance tuning. Your work will address core supply chain problems where machine learning delivers measurable business value.
On any given week, you might be designing a new feature engineering approach, running experiments to evaluate alternative modeling techniques, debugging model drift for a specific customer, or building pipeline infrastructure to operationalize a new ML capability. You will collaborate closely with product managers, professional services, software engineers, and customer-facing teams to translate complex supply chain challenges into well-scoped ML solutions.
This is a hands-on IC role with high autonomy and direct impact on customer outcomes and revenue. You will own ML projects end-to-end—the science and the engineering.
A Day in the Life
Research, design, and develop machine learning models for supply chain applications that drive measurable improvements in operational efficiency and planning accuracy
Perform exploratory data analysis, statistical modeling, and feature engineering on large, complex supply chain datasets to identify signals and improve model performance
Design and run experiments to evaluate new modeling approaches, loss functions, feature sets, and hyperparameter configurations—interpreting results and translating findings into production improvements
Build and maintain robust ML pipelines that process, clean, and transform data from enterprise supply chain systems (SQL databases, APIs, ERP integrations)
Deploy and maintain models in cloud-based production environments, managing the full lifecycle from training through inference and monitoring
Implement model evaluation, drift detection, and monitoring frameworks to ensure reliability across diverse customer environments
Diagnose and resolve model performance issues for individual customers—investigating data quality, feature behavior, and distributional shifts
Partner with product managers, professional services, and engineering teams to understand customer problems and scope ML solutions appropriately
Communicate findings, model behavior, trade-offs, and recommendations clearly to both technical and non-technical stakeholders
Contribute to the team’s technical direction on ML methodology, architecture, tooling, and best practices
Required Qualifications
Bachelor’s degree in Computer Science, Statistics, Data Science, Engineering, Operations Research, or a related technical field; or equivalent professional experience
3+ years hands-on experience in applied machine learning or data science roles, with models developed and deployed to production
Strong Python skills with experience writing clean, maintainable, production-grade ML code
3+ years professional SQL experience, including complex queries against large enterprise datasets
Deep understanding of statistical and machine learning methods: gradient boosting (LightGBM, XGBoost, CatBoost), regression, decision trees, clustering, time series techniques, and model evaluation methodology
Experience with feature engineering for structured and tabular data, including domain-informed feature design, temporal feature construction, and feature selection techniques
Demonstrated ability to design experiments, evaluate model performance rigorously, and iterate on approaches based on empirical results
Experience building and maintaining ML pipelines—data ingestion, feature engineering, training, evaluation, deployment
Working knowledge of cloud-based ML infrastructure (Azure preferred; AWS or GCP acceptable)
Strong communication skills with the ability to explain model behavior, experimental results, and trade-offs to non-technical audiences
Self-directed with a track record of owning ML projects end-to-end—from problem formulation through production delivery—with minimal supervision
Preferred Qualifications
Master’s or PhD in Computer Science, Statistics, Data Science, Engineering, Operations Research, or a related technical field
Experience in supply chain, operations, or logistics domains
Background in time series modeling, probabilistic methods, or optimization techniques applied to operational problems
Familiarity with Databricks, Spark, or similar distributed compute platforms for ML workloads
Experience with Azure services: Azure ML, Container Apps, App Configuration, DevOps pipelines
Experience working directly with enterprise customers to tune, validate, and explain model outputs in their specific business context
Experience with MLflow for experiment tracking and model versioning
Experience with Kafka or similar event streaming platforms for real-time data integration
Curiosity about the business processes your models serve and motivation to understand how supply chain decisions are actually made
Core Competencies
Customer Impact: Builds solutions with the end customer in mind—measures success by business outcomes, not model metrics alone
Analytical Depth: Goes beyond surface-level results to understand why models behave the way they do, especially when they fail—combines scientific rigor with practical problem-solving
Engineering Rigor: Writes production-quality code, designs reliable pipelines, and thinks about failure modes before they happen
Manages Complexity: Navigates messy real-world data and ambiguous problem definitions to deliver practical, scalable solutions
Communicates Effectively: Translates technical model behavior and experimental findings into clear narratives for product, services, and leadership audiences
Drives Results: Takes ownership, follows through on commitments, and delivers measurable improvements to customer outcomes
Technology Environment
Python, LightGBM, SQL, Azure (Container Apps, ML, DevOps), Databricks, Git/GitHub. Enterprise supply chain platform with SQL Server backends and REST APIs.
Why GAINS
- Work on software that leverages AI and ML to solve real logistics challenges for customers
- Direct impact on developer experience across the entire engineering org
- Collaborative, low-bureaucracy environment where engineers own their work end-to-end
- Competitive compensation and benefits