Senior Software Engineer, ML Platform

Senior Software Engineer, ML PlatformLocation: San Francisco, CA / Remote Flexible Employment Type: Full-time Focus: ML Platform, MLOps, Model Serving, Feature Stores, Underwriting InfrastructureAbout Our ClientOur client is building financial infrastructure that helps small businesses access the capital and products they need to grow.Their platform uses data, machine learning, and modern underwriting systems to power financial products at scale. As the company continues to expand, the infrastructure behind model experimentation, training, evaluation, inference, and retraining is becoming increasingly critical.This is an opportunity to join a high-impact infrastructure team and own the ML platform that enables data scientists to safely and quickly ship high-quality models into production.About the RoleOur client is hiring a Senior Software Engineer, ML Platform to lead the evolution of its machine learning platform.This person will design, build, and maintain the core systems that support model development, production deployment, batch inference, real-time inference, feature stores, observability, and underwriting infrastructure.You'll work closely with Data Science and Platform Engineering to turn research workflows into reliable software systems. This is a strong fit for an engineer who enjoys building developer-friendly platforms, creating clean abstractions, and owning infrastructure that powers real business decisions.What You'll DoOwn and evolve the company's ML platform end-to-endTurn data science notebooks into reusable, tested, production-ready software componentsBuild libraries, pipelines, templates, SDKs, and CLIs that help data scientists move fasterCreate developer-friendly abstractions for feature definition, model training, evaluation, deployment, and monitoringBuild and scale low-latency real-time model serving infrastructureExpand batch ML inference systems across scheduling, parallelism, cost controls, observability, failure handling, and rollbackOwn and improve the feature store, including offline and online feature definitionsDesign systems for high read/write throughput and consistent offline/online semanticsInstrument training and inference workflows for latency, throughput, accuracy, drift, data quality, and costBuild alerting, dashboards, and observability systems for platform healthSupport production underwriting systems across batch and real-time workflowsPartner with Data Science on model interfaces, SLAs, safety checks, and product integrationsDrive incident response, postmortems, and long-term reliability improvementsWhat We're Looking For5+ years of software engineering experienceExperience building ML platform, MLOps, model training, model deployment, or feature pipeline systemsStrong Python experienceStrong software design, testing, and platform engineering fundamentalsProficiency with SQLHands-on experience with Spark or PySparkStrong understanding of ML fundamentals, including probability, statistics, supervised and unsupervised learning, feature engineering, validation strategies, model evaluation, drift, stability, and monitoringExperience with modern data and ML infrastructure such as AWS, Databricks, MLflow, model registries, model serving, Airflow, or similar orchestration toolsExperience building real-time systems, including service design, caching, rate limiting, backpressure, and low-latency architectureExperience building batch pipelines at scalePractical knowledge of feature store concepts, including offline and online stores, backfills, point-in-time correctness, experiment tracking, and evaluation frameworksStrong ownership mindset and proactive approach to platform reliabilityExcellent communication and collaboration skills across engineering and data science teamsBonus ExperienceDeep Databricks experience, including MLflow, workflows, lakehouse architecture, or model servingExperience with feature stores such as Tecton, Feast, or similar platformsExperience with streaming technologies such as Kafka or KinesisExperience in fintech, risk, lending, underwriting, or regulated financial systemsFamiliarity with model safety checks, rejection flows, override flows, and auditabilityExperience with A/B testing platforms, shadow deployments, canary releases, and automated rollbackExperience building low-latency inference systemsWhy This OpportunityOwn a critical ML platform that powers underwriting and other ML-driven productsBuild infrastructure that helps data scientists ship models safely and quicklyWork across real-time inference, batch inference, feature stores, model evaluation, and platform observabilityPartner closely with Data Science and Platform Engineering on high-impact systemsBuild developer-friendly tools that create leverage across the technical organizationWork on meaningful infrastructure tied directly to financial access for small businessesStep into a senior role with end-to-end ownership over core ML platform systems