ML Engineer - Scikit-Learn & LightGBM, Scale & MLOps

Energy Innovation Corp; Charlottesville (VA); On-site; USD 84,000 - 130,000; Full time

Benefits offered by this job

Benefits and flexibility

Job summary

Grant Thornton is hiring a backend engineer to build scalable statistical modeling features, including Hypothesis Testing, in Charlottesville. You will scale services from a Charlottesville pilot to a Virginia-wide rollout, sketch sequence diagrams, and optimize performance for low latency and efficient resource use.

You will work with Scikit-learn and LightGBM, wrangle MLOps across environments, and help bridge MLOps with modeling tools to ensure smooth, end-to-end data flows.

Qualifications

• 3+ years applying statistical modeling in a technology setting.
• Hands-on proficiency with stress management and stakeholder management.
• Able to teach others and share knowledge effectively.

Responsibilities

• Ship the outcome-focused Statistical Modeling features.
• Scale services from Charlottesville pilot to VA-wide rollout.
• Sketch LightGBM sequence diagrams to clarify data flow.
• Optimize application performance, latency, and resource use at scale.
• Architect fault-tolerant distributed systems leveraging Statistical Modeling and Scikit-learn.
• Wrangle MLOps config across environments so Charlottesville staging mirrors production.
• Bridge MLOps and LightGBM so the platform components communicate.

Skills

Stress management

Stakeholder management

Interpersonal skills

Statistical modeling

Mentoring / teaching

Scikit-learn

LightGBM

MLOps

LangChain

Tools

Scikit-learn

LightGBM

MLOps

LangChain

Job description

Grant Thornton is hiring a backend engineer to build scalable statistical modeling features, including Hypothesis Testing, in Charlottesville. You will scale services from a Charlottesville pilot to a Virginia-wide rollout, sketch sequence diagrams, and optimize performance for low latency and efficient resource use.

You will work with Scikit-learn and LightGBM, wrangle MLOps across environments, and help bridge MLOps with modeling tools to ensure smooth, end-to-end data flows.