Senior Machine Learning Engineer

FINNY•New York, NY

84d•$200,000 - $230,000•Onsite

About The Position

Being a Machine Learning Engineer at FINNY means owning the models that power search, matching, ranking, recommendations, data and intelligent automation across the product. This is a model first role. While you’ll work with real production data and pipelines, your primary impact comes from designing, training, evaluating, and improving ML systems that directly shape user outcomes. You’ll partner closely with product, backend, and frontend teams to turn ambiguous problems into measurable model improvements.

Requirements

You’re a model builder at heart
You care deeply about how models learn, not just how pipelines run
You’re comfortable reasoning about loss functions, tradeoffs, and evaluation
You enjoy designing solutions when the problem is underspecified and data is imperfect
Very strong Python with extensive hands-on experience building ML systems.
Strong statistical and mathematical foundations.
Proven experience training, fine-tuning, and deploying custom models into production, not just experimentation or offline research
Experience designing loss functions, evaluation metrics, and validation strategies aligned with real-world product objectives
Familiarity with model lifecycle management: versioning, reproducibility, monitoring, and iteration in production environments
You’ve taken models beyond notebooks and into real products
You understand failure modes, monitoring, and iteration in production ML
Startup experience
You tackle ambiguity head-on and turn fuzzy problems into concrete experiments
You move fast, iterate, and aren’t precious about first approaches
You communicate clearly about model behavior, limitations, and tradeoffs

Responsibilities

Help build FINNY’s core models
Design, train, and iterate on custom models that power data imputation, prospect and audience recommendations, campaign customization and personalization, and automations.
Build & improve models in production
Take models from research → experimentation → deployment → iteration
Own offline evaluation, online metrics, and feedback loops
Improve model performance over time through better objectives, features, and training strategies—not just more data
Advanced modeling & experimentation
Apply and adapt techniques such as: Fine-tuning, RL methods (DPO), Transfer learning and weak supervision, Synthetic data generation and augmentation
Operate effectively in low-signal, noisy, or cold-start environments
Contribute to ML systems & infrastructure
Work with backend engineers to productionize models reliably and at scale
Help define standards for model versioning, evaluation, deployment, and monitoring
Influence long-term ML strategy and reduce technical debt in modeling workflows