Sr. Applied Machine Learning Engineer

Plug Technologies Inc•Los Angeles, CA

2d•$164,000 - $200,000•Onsite

About The Position

Plug is an automotive fintech startup building an electric-vehicle-only marketplace for consumers, dealers, and commercial partners to trade, sell, and auction used EVs. The company’s models support instant offers, auction pricing, battery state-of-health analysis, and related marketplace decisions. The Sr. Applied Machine Learning Engineer will join the Intelligence team and work with the ML engineer and data team to improve and extend these models, including price and volume forecasts, battery models built from telematics, and LLM pipelines that extract structured information from vehicle, title, and inspection documents. The role covers the full machine-learning lifecycle, from feature pipelines, training, and evaluation through deployment, monitoring, drift detection, and retraining. Models are evaluated both statistically and against business outcomes such as offer acceptance, margin, and arbitration rates. The engineer will also help strengthen team practices for experiment tracking, model management, reproducible training, and ML code review. This is a full-time, in-person position at Plug’s Sawtelle office in Los Angeles. Relocation assistance is available for candidates who need to move; visa sponsorship is not available.

Requirements

At least 5 years of experience building and shipping machine-learning systems, including supporting multiple production models after launch.
Strong Python skills and deep experience with at least one modern ML framework, such as PyTorch, scikit-learn, XGBoost, or LightGBM.
Practical experience with forecasting, regression, or pricing problems involving tabular and time-series data.
Strong grounding in statistics and experiment design, including validation strategy, data leakage, calibration, and metric selection.
Comfort working with messy real-world data and strong SQL skills.
Hands-on experience building with LLMs, including tool use, retrieval, structured output, and evaluation.
Experience building personal tooling around coding agents, such as skills, MCP servers, custom workflows, or evaluations.
Habit of documenting work through design documents, model cards, and runbooks.
Enjoy working in a small team through pairing, reviewing colleagues’ work, and sharing credit.
Comfort working within systems designed by others, making a case when disagreeing, and then shipping the agreed plan.
High agency in an ambiguous, fast-moving startup environment, including finding valuable problems and shipping solutions without waiting for a perfect specification.
BS or MS in a quantitative field such as computer science, statistics, engineering, physics, or applied mathematics, or equivalent experience.

Nice To Haves

Experience with vehicle, marketplace, auction, lending, or insurance pricing data.
Experience with sensor or telematics data, battery modeling, or other physical-systems data.
Experience with transfer learning, multi-task learning, or model distillation.
Experience with ML or LLM serving under latency and cost constraints.
Experience with a modern data stack such as Snowflake or dbt.

Responsibilities

Build, improve, and run in production models for instant offers, auction pricing, sell-through, valuation, price and volume forecasting, and demand signals.
Develop battery state-of-health and vehicle-condition models using telematics and other time-series data, including data that is noisy, sparse, or partly labeled.
Work with the team across the full model lifecycle, including feature pipelines on the Snowflake/dbt stack, training, evaluation, deployment, monitoring, drift detection, and retraining.
Evaluate models using statistical accuracy and business outcomes such as offer acceptance, margin, and arbitration rate, and help build feedback loops connecting those measures.
Use LLMs where appropriate to extract structured data from vehicle, title, and inspection documents, and build evaluations to assess their reliability.
Strengthen team practices for experiment tracking, model registry, reproducible training, and ML code review.
Share knowledge, pair on challenging problems, and learn from teammates with deep knowledge of the company’s data.