ADAS Analytics ML Scientist – Vision, LLMs & Retrieval

Mercedes-Benz Group; San Jose (CA); On-site; USD 118,000 - 148,000; Full time

Benefits offered by this job

Medical, dental, vision insurance

401(k) with employer match

Up to 15 company-paid holidays

Paid time off and parental leave

Tuition assistance program

Wellness reimbursement

Job summary

Mercedes-Benz Research and Development North America (MBRDNA) is seeking a Data Scientist to scale ADAS analytics and build vision-language models for SSR video and fleet-scale insights. You will develop embedding-based retrieval, set up LLM classification, and provide human-readable summaries for management and engineers.

You will deploy PySpark/Delta Lake pipelines on Azure, quantify firmware impacts on KPIs, and prototype ranking and anomaly-detection methods.

Qualifications

• Bachelor's or Master's in a quantitative field; advanced degrees beneficial but not required.
• 3+ years of professional experience.
• 2-5 years of experience in data science or applied machine learning.
• Depth in at least one: foundation models, embeddings, or production ML.
• Strong Python (NumPy, pandas, scikit-learn) and SQL.
• Ability to communicate findings to engineers and management.

Responsibilities

• Apply vision-language models to SSR video and multimodal data.
• Build LLM classification and reasoning pipelines for event summaries.
• Design embedding pipelines and semantic search for similar-event retrieval; cluster at fleet scale.
• Develop model-based ranking and active-learning loops.
• Quantify firmware updates impact on KPIs; segment driver cohorts.
• Deploy models into PySpark/Delta Lake pipelines and FastAPI analytics API on Azure.

Skills

Python

SQL

LLMs / VLMs

PySpark

PyTorch

TensorFlow

Education

Bachelor's or Master's in Data Science/CS/Statistics

Tools

Azure data platform

Delta Lake

FastAPI

MLflow

Weights & Biases

FAISS

Pinecone

pgvector

FAISS

Job description

Mercedes-Benz Research and Development North America (MBRDNA) is seeking a Data Scientist to scale ADAS analytics and build vision-language models for SSR video and fleet-scale insights. You will develop embedding-based retrieval, set up LLM classification, and provide human-readable summaries for management and engineers.

You will deploy PySpark/Delta Lake pipelines on Azure, quantify firmware impacts on KPIs, and prototype ranking and anomaly-detection methods.