Machine Learning Scientist

Harnham

United States

Remote

USD 140,000 - 190,000

Full time

30 hours ago

Be an early applicant

Get a reply from this employer

Your resume and cover letter, tailored to exactly what they’re hiring for.

Get past ATS filters

Benefits offered by this job

Annual bonus

Equity

Unlimited PTO

401(k) matching

Medical, dental, and vision

Job summary

Harnham is hiring Machine Learning Scientists and Senior Machine Learning Scientists for a fully remote (U.S.-based) role with occasional on-site team gatherings. You will develop original ML solutions for fraud/risk and transaction decisioning, owning experiments from hypothesis to production deployment.

The team emphasizes rigorous evaluation and end-to-end ownership, with opportunities for staff-level work.

Qualifications

4–6+ years of experience in production-grade machine learning environments.
Strong Python, SQL, and Spark/PySpark experience, including hands-on work with large-scale or distributed ML/data processing.
Original ML development: personally building ML solutions from initial business problems.
Scientific experimentation: designing experiments, comparing approaches, selecting progress.
Strong ML foundations: theory, evaluation, experimentation, performance measurement.
End-to-end ML ownership: from experimentation to production deployment and monitoring.
Independent technical ownership: evaluating approaches and solving unspecified problems.
Production-scale ML experience: creating models/workflows for large-scale production.

Responsibilities

Develop original ML solutions and translate business problems into hypotheses and production-ready models.
Own scientific experimentation: design and run experiments on architectures, features, training, and decisioning schemes.
Evaluate model performance with rigorous statistical methodologies.
Build production ML systems and oversee deployment, monitoring, and iteration.
Work at scale using Python, SQL, and Spark/PySpark in distributed environments.
Improve ML infrastructure and tooling for efficient experimentation.
Collaborate with Engineering, Product, and Risk teams to align modeling with business goals.
Drive technical decisions and contribute to standards.

Skills

Python

SQL

Spark/PySpark

ML development

Experimentation

End-to-end ownership

Production ML

Independent ownership

Education

PhD in CS/Stats/ML

Master’s in quantitative field

Job description

Machine Learning Scientist / Senior Machine Learning Scientist

Location: United States (Remote)

Work Arrangement: Fully Remote, with approximately one team onsite gathering per year

Base Salary: $140,000–$190,000 depending on location and experience, with flexibility up to approximately $210,000 for select senior-level candidates

Work Authorization: U.S. Citizens and Green Card holders only. Visa sponsorship is not available.

About the Company

We're partnering with an established, Series E fintech company that uses advanced machine learning to help e-commerce businesses prevent fraud, reduce financial losses, and improve transaction approval rates.

Its AI-powered platform analyzes online transactions in real time, helping merchants distinguish legitimate customers from fraudulent activity while providing financial protection against fraud losses.

The company supports approximately 1,000 merchants across more than 100 countries, operating at significant transaction volume and scale.

Machine learning is central to the company's product and business model. The organization continues to invest in its ML capabilities, infrastructure, and scientific experimentation to improve fraud detection and transaction decisioning.

About the Role

We're hiring Machine Learning Scientists and Senior Machine Learning Scientists, with flexibility for Staff-level candidates, to join a highly technical applied machine learning organization.

This is a hands-on individual contributor opportunity focused on developing original machine learning solutions for complex fraud, risk, and transaction decisioning problems.

The team is looking for scientists who can take an ambiguous business problem, formulate hypotheses, develop and compare modeling approaches, rigorously evaluate results, and bring successful solutions into production.

Unlike roles primarily focused on deploying existing models or integrating AI APIs, this position emphasizes original ML development, scientific experimentation, statistical evaluation, and end-to-end production ownership.

You’ll work within a collaborative ML organization of approximately 20 people, organized into smaller teams that own their models and services end-to-end.

Examples of upcoming projects include:

Ensemble Model Architecture: Developing a new modeling architecture that combines multiple fraud detection models into an improved decision-making system.
Offline Model Evaluation: Building and improving methodologies for evaluating model performance before production, including how fraud outcomes are identified and labeled during evaluation.

These are representative initiatives, with opportunities to contribute to other areas of large-scale ML development and experimentation.

What You’ll Do

Develop original ML solutions: Translate complex business problems into hypotheses, modeling approaches, experiments, and production-ready machine learning solutions.
Own scientific experimentation: Design and execute experiments involving model architectures, feature engineering, training methodologies, data, labels, and decisioning strategies.
Evaluate model performance: Apply statistical and experimental methodologies to compare approaches, validate improvements, and determine which models should advance toward production.
Build production ML systems: Take successful models from experimentation through deployment, monitoring, performance measurement, and continued iteration.
Work at scale: Develop and optimize machine learning workloads using Python, SQL, and Spark/PySpark within large-scale distributed computing environments.
Improve ML infrastructure: Develop and enhance ML pipeline components, tools, and processes that support efficient experimentation and scalable model development.
Collaborate across teams: Partner with Engineering, Product, and Risk teams to connect modeling decisions with business objectives and measurable outcomes.
Drive technical decisions: Independently evaluate modeling approaches, identify tradeoffs, and contribute to technical direction and development standards.

What We’re Looking For

4–6+ years of experience working in production-grade machine learning environments.
Strong Python, SQL, and Spark/PySpark experience, including hands‑on work with large-scale or distributed ML/data processing.
Original machine learning development: Experience personally building ML solutions from initial business problems rather than primarily implementing, deploying, or fine‑tuning predefined models.
Scientific experimentation: Demonstrated experience developing hypotheses, designing experiments, comparing modeling approaches, and determining which solutions should progress.
Strong ML and statistical foundations: Solid understanding of machine learning theory, statistical evaluation, experimental methodology, and model performance measurement.
End-to-end ML ownership: Experience taking machine learning solutions from experimentation through production deployment, monitoring, and iteration.
Independent technical ownership: Ability to evaluate competing approaches, make modeling decisions, and solve problems where the solution is not predefined.
Production-scale ML experience: Experience developing models and ML workflows that operate reliably in large-scale production environments.

We’re particularly interested in candidates whose backgrounds are closer to Applied Scientist, Machine Learning Scientist, or research-oriented Data Scientist roles, with substantial hands‑on modeling and experimentation experience.

Nice to Have

Experience with fraud detection, payments, financial risk, transaction decisioning, or related quantitative decisioning problems.
Direct experience developing classification models, particularly for fraud, abuse, or other adversarial modeling problems.
Experience designing ensemble models or sophisticated offline model‑evaluation frameworks.
PhD in Computer Science, Statistics, Machine Learning, or another highly quantitative discipline. Strong Master’s‑level candidates are also welcome.
Research‑intensive backgrounds involving statistical experimentation, advanced predictive modeling, or applied machine learning research.

Why This Role?

Machine learning is central to the business. Your models will directly influence how a large‑scale platform evaluates transactions, identifies fraud, and supports legitimate customer approvals.

Significant technical ownership. Scientists are trusted to investigate complex problems, develop their own hypotheses, select modeling approaches, and influence technical direction rather than simply execute predefined tasks.

A strong scientific environment. The team emphasizes rigorous experimentation, statistical evaluation, and continuous learning. Successful experiments can progress into production, while unsuccessful approaches provide valuable information for future development.

Challenging, high‑impact projects. Work on sophisticated ML problems involving ensemble architectures, classification, offline evaluation, and large‑scale transaction decisioning.

End-to-end development. You’ll have the opportunity to work across the complete ML lifecycle, from initial research and experimentation to production systems and measurable business outcomes.

Established ML infrastructure. Join a team with significant investment in internal ML tooling, distributed computing, and scalable experimentation capabilities.

Compensation and benefits include: