Machine Learning for Supply Chain

Samsara

Location not specified

Full Time

Posted April 17, 2026

$170k - $257k

Not Specified

~78 people viewed this recently

Apply Now

Application opens on company website

Job Description

This is a senior AI/ML role focused on developing and deploying machine learning models to optimize Samsara's global supply chain operations, including demand forecasting, inventory management, and supplier risk assessment, leveraging large-scale real-time IoT and enterprise data.

Key Responsibilities

Design, develop, and deploy machine learning models for demand forecasting, inventory optimization, and supplier risk scoring.
Build data pipelines and ETL processes using ERP, IoT, and third-party datasets to support modeling efforts.
Implement MLOps best practices for model training, deployment, monitoring, and performance tracking.
Collaborate with cross-functional teams to align data requirements, integrate models, and support operational decision-making.
Lead initiatives to improve data infrastructure, analytics platforms, and real-time inference capabilities.
Mentor junior data scientists and contribute to scientific strategy, experimentation frameworks, and modeling standards.
Establish governance, documentation, and quality controls for the entire model lifecycle.
Drive adoption of AI tools within supply chain teams and train staff on insights-driven workflows.
Define the AI transformation roadmap for the supply chain in collaboration with executive leadership.

Requirements

8 years in applied data science or machine learning, ideally in supply chain, operations research, logistics, or manufacturing.
Master's degree or PhD in Computer Science, Statistics, Data Science, Electrical Engineering, Operations Research, or related technical field.
Expertise in statistical modeling and machine learning techniques including time series forecasting, optimization, anomaly detection, and causal inference.
Strong Python coding skills and fluency in SQL, with proven experience developing and deploying production machine learning systems.
Proficiency in MLOps practices including automated testing, CI/CD, model versioning, monitoring, and performance tracking.
Experience building real-time inference pipelines and managing GPU/TPU resources for training at scale.
Familiarity with data visualization tools such as Tableau and Power BI, and cloud platforms including AWS, GCP, or Azure.
Ability to design and validate experiments such as A/B tests, multi-armed bandits, and Bayesian methods for uncertainty quantification.
Proven track record building advanced forecasting models like Prophet, LSTMs, or Transformer-based models for intermittent demand and seasonal patterns.
Skilled in profiling ML workloads and reducing cloud GPU spend through model compression, quantization, pruning, and serverless architectures.
Demonstrated ability to deploy low-latency inference systems across multiple geographic regions with fail-over and disaster-recovery strategies.
Experience evaluating and integrating third-party ML platforms and contributing to or leveraging open-source projects.
Deep understanding of supply chain concepts such as S&OP, IBP, safety stock, EOQ, ERP systems like SAP, NetSuite, E2DP, Propel, and inventory optimization theory.
Ability to work with large, noisy, high-dimensional datasets from ERP data, IoT sensor streams, third-party market intelligence, and operational metrics.
Experience in building features from real-world signals such as vehicle diagnostics, utilization patterns, and failure modes.
Experience in designing and validating experiments, including A/B testing, multi-armed bandits, and Bayesian methods.
Ability to build and deploy models for demand forecasting, lead times, cellular spend prediction, and supply chain risk modeling.
Experience in creating pipelines and ETL jobs to serve models and stakeholders using large-scale ERP, IoT, and third-party datasets.
Ability to deliver production-grade code supporting batch and real-time inference, following MLOps best practices.
Partnering with cross-functional teams including Product, Engineering, Procurement, and Finance to align on data requirements and integration.
Mentoring junior scientists through code reviews and collaborative projects.
Establishing governance frameworks, documentation standards, and quality controls for model development, validation, and lifecycle management.
Leading initiatives to improve data infrastructure and analytics platforms for real-time model training, monitoring, and inference at scale.
Hiring, developing, and leading an inclusive, high-performing team aligned with company cultural principles.
Ability to explain complex technical concepts to non-technical stakeholders and drive adoption of data-driven solutions.