Principal Machine Learning Engineer

Work at the intersection of applied ML and platform engineering. Collaborate with Research Scientists, Data Scientists, and ML Platform Engineers. Design tools and systems to accelerate model development and improve predictive accuracy. Lead engineering initiatives to build scalable, reusable infrastructure. Build a unified embeddings platform for training, serving, and managing representations at scale; streamline feature engineering pipelines to reduce manual steps and deliver new signals quickly; develop automated continuous-learning systems for data refresh, retraining, evaluation, and drift monitoring with minimal manual effort; and scale our training pipelines to support larger datasets, more complex architectures, and faster experimentation. Work backward from applied ML projects to harden platform capabilities for ML teams. Define roadmap for next generation ML Platform balancing near-term impact with long-term scalability. Collaborate cross-functionally with Data Engineering, ML Platform, Pricing, and other teams. Multiply effectiveness of every ML team at Upstart by accelerating innovation. Combine science innovation with cross-functional collaboration and advisory.

Minimum Qualifications

• 7+ years hands-on experience in applied machine learning with production-scale modeling.
• Expertise in end-to-end model development: data prep, feature engineering, training, evaluation, deployment.
• Experience in high-scale ML-driven product environments (fintech, pricing, risk modeling preferred).
• Proficiency in Python and core ML frameworks (PyTorch, TensorFlow, Scikit-learn, XGBoost).
• Ability to work autonomously and lead technical direction in ambiguous domains.
• Experience collaborating with cross-functional teams including scientists and engineers.
• Ability to bridge engineering and science teams; influence technical strategy across disciplines.
• Numerically savvy; able to operate at fast pace.
• Master’s degree or PhD in quantitative discipline or equivalent experience.

Preferred Qualifications

• Experience optimizing ML workflows using CUDA/GPU acceleration.
• Background in feature store design, embedding architecture, synthetic data generation for training.
• Proven track record improving model accuracy in production with measurable business outcomes.
• Familiarity with experimentation frameworks, hyperparameter tuning tools, automated model selection.

Location

United States | Remote work available Offices located in Columbus (OH), Austin (TX), Bay Area (CA), New York City (opening Summer 2026) Digital-first company allowing flexibility to work remotely or from office locations.

Employment Type

Not explicitly stated; implied full-time role.

Remote Work Info

Yes. Upstart is digital-first offering flexibility to work remotely across the US or from offices.

Experience Level

Senior level - Principal Engineer role requiring 7+ years experience.

Skills Mentioned

• Applied machine learning
• Model development lifecycle (data prep, feature engineering, training, evaluation, deployment)
• Python programming
• ML frameworks: PyTorch, TensorFlow, Scikit-learn, XGBoost
• CUDA/GPU acceleration (preferred)
• Feature store design
• Embedding architecture
• Synthetic data generation
• Experimentation frameworks
• Hyperparameter tuning
• Automated model selection

Compensation

Anticipated base salary range: $220700—$300000 USD depending on location. Additional bonuses and equity compensation offered.

Benefits Highlights

Competitive compensation; 401(k) matching; employee stock purchase plan; medical/dental/vision coverage; paid time off; family leave; mental health support; financial wellness resources; annual wellness/productivity allowances; onsite perks at offices. Equal Opportunity Employer statement included.