Senior Data Scientist

Phaedon•Minneapolis, MN

7d•$105,000 - $155,000•Hybrid

About The Position

We are looking for a Senior Data Scientist who is a builder, not just a maintainer. This is a high-ownership opportunity for an AI/ML engineer who wants to design and ship the models that power our loyalty platform in production, not just prototype them. You'll build AI/ML capabilities that our SaaS product calls at runtime: fraud detection, personalization, recommendation, and forecasting models served through APIs, not one-off notebooks handed to someone else to productionize. We're looking for a self-starter who identifies opportunities to apply AI/ML to the product roadmap, proposes the approach, builds it, ships it, and owns it in production. The primary focus of this role is product-embedded model development. There will be some client-facing work; however, it is anticipated to be a small portion of the role.

Requirements

Bachelor's degree in data science, computer science, computer engineering, or related field AND 5+ years of hands-on experience building and shipping ML models into production systems OR equivalent combination of education and experience
Demonstrated track record of taking a model from idea to production-serving endpoint inside a live product, not just research/POC work; be prepared to speak to specific systems you built that are running in production today
Fluency in the full model lifecycle: data/feature engineering, training, evaluation, deployment, versioning, monitoring, and retraining
Knowledge of Infrastructure-as-Code (Terraform, Pulumi, CloudFormation) for deploying ML infrastructure repeatably
Experience with source control and automated deployment pipelines (Git, Docker)
A demonstrated self-starter mindset: comfortable identifying a product opportunity, scoping the technical approach, and driving it to completion with minimal guidance
Strong written and verbal communication skills to document and present technical approaches to engineering and product stakeholders
Advanced Python (including AI/ML libraries like transformers, LangChain), SQL, Boto3
AWS Bedrock, SageMaker, prompt engineering, model fine-tuning
AWS services, particularly Bedrock, SageMaker, Lambda, Redshift, Athena, and Glue
Experience with Superset, Tableau, and/or Power BI
Object-oriented programming, testing frameworks, CI/CD, model versioning

Nice To Haves

Direct experience building models that are embedded in and called by a live SaaS product (recommendation engines, fraud/anomaly detection, personalization, forecasting, chatbots)
Experience with vector databases and RAG implementations in production
Knowledge of LLM fine-tuning, evaluation, and deployment strategies at scale
Strong MLOps background: model versioning, automated retraining, drift detection, canary/shadow deployments
Experience with API development and microservices architecture in a product engineering context
Background in fraud detection, loyalty/rewards platforms, or marketing/AdTech modeling a plus
Prior experience balancing product engineering with occasional client-facing technical work

Responsibilities

Design, build, and own AI/ML models that are directly integrated into and called by our SaaS product in production
Own the full model lifecycle: problem framing, data/feature design, training, evaluation, deployment as a callable service, and post-deploy monitoring/retraining
Build and maintain production inference APIs and microservices that serve model predictions to the product with defined latency and reliability SLAs
Implement and productionize models using AWS Bedrock, SageMaker, and other AWS AI services, going beyond POC into hardened, versioned, production systems
Develop RAG (Retrieval-Augmented Generation) systems and other LLM-powered features as first-class product capabilities
Proactively identify where AI/ML can create product differentiation (fraud detection, member behavior prediction, personalization/recommendation, anomaly detection) and bring proposals forward rather than waiting for requirements to be handed down
Build and manage SageMaker training pipelines, model registry, and endpoint deployments, including feature store integration and automated retraining triggers
Build automation, monitoring, and alerting for production ML systems using Lambda and other AWS services
Create and maintain Infrastructure-as-Code (Terraform, Pulumi, CloudFormation) for all model and pipeline infrastructure, no manual, undocumented deployments
Build data pipelines that synthesize complex datasets from multiple sources into model-ready features
Develop CI/CD pipelines for automated deployment and model versioning; implement model registry and rollback practices
Implement error-proofing, integration testing, and monitoring/logging for AI systems running in production
Support select client engagements where deep technical model expertise is needed to scope or validate an AI/ML approach
Partner with product and analytics leadership to translate roadmap priorities into shipped model capabilities
When client-facing, present technical findings and recommendations with clarity to both technical and business stakeholders