Senior Software Engineer - Machine Learning Platform

The Team:

The Machine Learning and Simulations Platform (MLSP) team builds and operates the core infrastructure that powers ML model training,feature engineering, inference, and marketplace simulation at Upstart. Every underwriting, fraud, conversion, and verification model runs on this platform. We own the full production path: the data and features that feed a model, the infrastructure that serves it at decision time, the tooling that deploys it, and the simulation systems that predict business impact before a change goes live.

We are reimagining that platform to keep pace with our ML teams. That work spans low latency and GPU model serving, self-service model deployment, a feature platform that gives ML one place to define and serve production features, and high fidelity marketplace simulation. The team partners closely with ML, Engineering, Product, Data Platform.

As a Senior Software Engineer on the ML and Simulations Platform team at Upstart, you will be responsible for building an MLOps platform to support machine learning model inference, process automation, model deployment, and observability. Machine Learning is critical to Upstart’s core business, and our greatest competitive advantage lies in the fact that we’re able to innovate on our AI engine quickly. You will also help build a marketplace simulation platform to support rapid innovation across ML and Finance teams.

How you’ll make an impact

• Build, maintain, and optimize Upstart’s next-generation machine learning and simulation platform, enabling increased scale, performance, and confidence in decisioning.
• Develop high-quality software applications that enable machine learning models to be applied to the ever-evolving needs of the business
• Build self-service tooling so ML teams can register features and deploy models independently, and reduce the manual work the platform team absorbs today.
• Deliver the data and feature infrastructure behind every model, including feature definition, storage, serving, and offline to online parity.
• Design and contribute to our simulation systems to more accurately reflect production environments, reducing simulation cost and enabling broader usage across teams.
• Communicate closely with cross-functional partners from ML, Engineering, Product, and Data Engineering teams, keeping all stakeholders informed
• Mentor engineers across the team, sharing expertise on distributed systems,MLOps, and scalable architecture.

Minimum Qualifications

• 6+ years of software engineering experience.
• Experience building and maintaining backend software services and APIs.
• Experience with distributed systems or large scale data processing, using Spark, Databricks, Ray, or an equivalent.
• Experience with an ML platform or the ML production path, such as training pipelines, model serving, feature pipelines, or a training data platform.
• Proficiency with some or many of the following: Python, Kotlin, Databricks, and AWS.
• Exhibits a growth mindset. You pick up new technologies that fit the task, and you learn from others.
• Ability to quickly comprehend complex requirements from ML, product, or engineering leadership, and translate them for both technical and non-technical partners.

Preferred Qualifications

• Skill with Metaflow, MLflow, gRPC, Spark/PySpark, dbt, Ray, GPU
• Knowledge of simulation, experimentation, or backtesting systems.
• Experience building self-serve or configuration driven tooling for internal users.
• Excellent quantitative reasoning skills with interest in working at the intersection of engineering and machine learning.
• Strong sense of ownership and accountability for the quality and timely delivery of work.
• Excellent written and verbal communication skills with partners, peers, and product owners.
• Ability to thrive in self-directed work and in collaborative settings, contributing positively to team dynamics.

Position location This role is available in the following locations: Remote

Time zone requirements The team operates on the East/West coast time zones.

Travel requirements As a digital first company, the majority of your work can be accomplished remotely. The majority of our employees can live and work anywhere in the U.S but are encouraged to to still spend high quality time in-person collaborating via regular onsites. The in-person sessions’ cadence varies depending on the team and role; most teams meet once or twice per quarter for 2-4 consecutive days at a time.

#LI-REMOTE

#LI-MidSenior