Staff Machine Learning Engineer

tilthq; Stati Uniti; Remoto; USD 150.000 - 190.000; Tempo pieno

Vantaggi offerti da questo lavoro

Equity

Home office stipend

Flexible health plans

In-person gatherings

Descrizione del lavoro

tilthq is building a platform layer that connects data science with infrastructure for production machine learning systems. You will own a managed feature store and improve training, deployment, and monitoring workflows so data scientists can develop models with less engineering support.

The platform serves multiple teams and models, with growing needs around deep learning and streaming workloads. The role emphasizes hands-on building and operating ML platforms, feature stores, and scalable

Competenze

• Deep hands-on experience building or operating software or platform systems for machine learning.
• Experience with feature stores and model deployment platforms; managed feature store experience is relevant.
• Strong ML CI/CD, model registry, and infrastructure-as-code experience, including Terraform.
• Cloud deployment experience with AWS or Azure, Kubernetes, workflow orchestration, and batch/real-time serving.
• Evidence of self-service tooling for data scientists and collaboration with backend engineers to productionize models.
• Ability to work through ambiguity, gather needs, propose solutions, and drive adoption.

Mansioni

• Run and improve the feature store, including tools for defining, backfilling, and serving features for training and inference.
• Build self-service training and deployment workflows for registering features, launching jobs, promoting models, and monitoring.
• Make training and retraining repeatable, schedulable, and grounded in shared feature data.
• Scale batch scoring and low-latency inference with attention to autoscaling, resource isolation, latency, and cost.
• Extend monitoring for data and feature drift, prediction quality, and serving health with dashboards and alerts.
• Set and encourage adoption of standards for model and feature versioning, lineage, testing, and documentation.
• Work directly with data scientists and backend engineers to identify pain points and deliver tools teams adopt.

Conoscenze

Feature store

ML platform

Terraform

AWS/Azure

CI/CD for ML

Kubernetes

Model registry

Data pipelines

Monitoring dashboards

Strumenti

Databricks

Descrizione del lavoro

Role overview

Build and operate the platform layer that connects data science with infrastructure for production machine learning systems. You'll own a managed feature store and improve training, deployment, and monitoring workflows so data scientists can develop and release models with less engineering support. The platform serves multiple teams and models, with growing needs around deep learning and streaming workloads.

Responsibilities

• Run and improve the feature store, including tools and patterns for defining, backfilling, and serving features for both training and inference.
• Build self-service training and deployment workflows that let data scientists register features, launch jobs, promote models, and set up monitoring.
• Make training and retraining repeatable, schedulable, and grounded in shared feature data.
• Scale batch scoring and low-latency inference with attention to autoscaling, resource isolation, latency, and cost.
• Extend monitoring for data and feature drift, prediction quality, and serving health, with useful dashboards and alerts.
• Set and encourage adoption of standards for model and feature versioning, lineage, testing, and documentation.
• Work directly with data scientists and backend engineers to identify workflow pain points and deliver tools that teams adopt.

Requirements

• Deep hands-on experience building or operating software or platform systems for machine learning, including platforms serving multiple production models and teams.
• Experience with feature stores and model deployment platforms; managed feature store experience is especially relevant.
• Strong machine learning CI/CD, model registry, and infrastructure-as-code experience, including Terraform.
• Cloud deployment experience with AWS or Azure, Kubernetes, workflow orchestration, and batch and real-time model serving.
• Evidence of building self-service tooling that data scientists use, and collaborating with backend engineers to put models into production.
• Ability to work through ambiguity, gather needs, propose solutions, and carry projects through to adoption.
• Enough knowledge of model development and common algorithms to work as a technical partner to data scientists.

Nice to have

• Familiarity with PySpark, Databricks, or model monitoring tools such as Evidently AI.

Benefits and work setup

• Remote-first collaboration across countries and time zones, with home office reimbursement and paid in-person gatherings twice yearly.
• Compensation includes equity; flexible health plans and subsidies are offered.