Forecasting ML Scientist: Time Series & Vertex AI

Valiant Technologies LLC; Lansing (MI); On-site; USD 120,000 - 180,000; Full time

Job summary

Valiant Technologies LLC is seeking a seasoned Data Scientist to own our machine learning sales forecasting end-to-end. You will design, build, and productionize forecasting models that guide planning, inventory, and revenue decisions using a modern Google Cloud stack (BigQuery, Vertex AI).

You’ll work with Sales, Finance, and Operations to translate business questions into forecasting problems, establish accuracy benchmarks, and drive continuous improvement across data quality and deployment

Qualifications

• 8+ years of experience as a Data Scientist or ML Engineer.
• Hands-on experience shipping models to production.
• Strong time series and forecasting expertise.
• Proficient in Python and BigQuery at scale.
• Experience with Vertex AI and MLOps.

Responsibilities

• Build, validate, and deploy ML-based sales forecasting models for short- and long-horizon predictions.
• Engineer features and training data directly in BigQuery with SQL transformations.
• Train and deploy models on Vertex AI with pipelines and endpoints.
• Set up model monitoring, retraining triggers, and drift tracking.
• Collaborate with Sales, Finance, and Operations to translate questions into forecasts.
• Define forecast accuracy benchmarks and improve them.
• Contribute to data quality and MLOps best practices.

Skills

Time series forecasting

Python

SQL

Model evaluation

Stakeholder communication

Tools

BigQuery

Vertex AI

BQML

TensorFlow

Prophet

Job description

Valiant Technologies LLC is seeking a seasoned Data Scientist to own our machine learning sales forecasting end-to-end. You will design, build, and productionize forecasting models that guide planning, inventory, and revenue decisions using a modern Google Cloud stack (BigQuery, Vertex AI).

You’ll work with Sales, Finance, and Operations to translate business questions into forecasting problems, establish accuracy benchmarks, and drive continuous improvement across data quality and deployment