Sr Data Scientist

Overview

Senior Data Scientist

Epicor Prism

You’ll collaborate with a cross-functional team of engineers, product managers, and fellow data scientists—driven by a shared focus on efficiency, experimentation, and customer impact. You’ll work closely with business stakeholders to gather feedback and continuously refine AI features that directly improve user experience and business outcomes.

What You'll Do

• Design and iterate on LLM-powered agents using Prism’s internal frameworks
• Build prompts and context strategies for tasks such as data summarization, workflow automation, image-text comparison, etc.
• Run structured, reproducible experiments—tracking versions, prompt variants, latency, and failure cases
• Collaborate with engineers and product managers to integrate agents into real Epicor systems
• Design telemetry to monitor agent performance, usage trends, and error patterns
• Contribute to internal best practices in prompting, evaluation, and context management

What We’re Looking For

• Experience:

  4+ years in data science or applied ML roles; 1+ year developing real applications using LLMs

• Prompting:

  Strong grasp of prompt engineering patterns, task decomposition, and few-shot strategies

• Context Management:

  Familiarity with managing long contexts, chunking, summarization, and retrieval (RAG)

• Experimentation:

  Proven experience in tracking experiments, evaluation design, metric creation, and telemetry

• Tools:

  Python, Git

• Collaboration:

  Excellent written and verbal communication; ability to explain tradeoffs and iterate quickly

Good to Have

• Experience with multimodal tasks (e.g., comparing product specs and images)
• Familiarity with model fine-tuning or RLHF
• Experience with MLOps platforms like ClearML, WandB, or Neptune
• Background in ERP, compliance, or supply chain
• Exposure to .NET/C# environments and lightweight deployment workflows (containers, APIs, CI/CD)

Success Metrics

• High-quality, reliable agents aligned with task goals
• Clean, versioned experimentation logs and telemetry to support rapid iteration
• Delivered features that produce measurable time or cost savings
• Reusable components (e.g., prompt patterns, eval methods) that help other Epicor teams move faster

This Role Is Ideal For

An LLM-savvy data scientist who wants to build real AI systems—not just demos. Someone who tests prompts like code, tracks experiments like products, and loves to ship. You’ll have autonomy from day one, meaningful projects to lead, and a platform that supports your technical growth.