Software Engineer, Sweeps (Remote)

Key Skills

Design, GraphQL, Machine learning techniques, Python Programming, Streaming APIs, TypeScript

Job Description

As the software engineer in charge of Weights & Biases Sweeps product, you’ll build and maintain key features to allow users to effectively tune hyperparameters. You’ll work on the next generation of sweeps, with a focus on allowing users to effectively distribute their hyperparameter tuning scripts across clusters and cloud infrastructure. You’ll also contribute to the development of Weights & Biases Launch, an execution layer being actively developed on the Weights & Biases platform. You will have the opportunity to design system architecture and APIs so users can easily send their workflows to the cloud, or local clusters.

Responsibilities

• Design and implement features to help users build and tune the best models by leveraging large scale compute.
• Lead the maintenance of our existing Sweeps product and ensure a seamless transition to the next generation.
• Contribute to the design and implementation to an execution platform built on Weights & Biases.
• Focus on our end users and listen to customer feedback in order to deliver value.

Requirements

• 4+ years of experience in software engineering
• Strong software engineering fundamentals and knowledge of at least 1 modern programming language (Python, Go, Typescript, GraphQL, etc)
• Strong empathy for ML practitioners with a keen interest in understanding and addressing their challenges.
• A desire to be self-driven and find the path to greatest impact.
• Nice to have: experience working with ML Engineers and building tools to allow them to easily connect with their compute.

Why join us?

• Top-tier machine learning teams rely on our tools for their daily work at companies including OpenAI, Toyota Research Institute, Lyft, Samsung, and Pandora.
• You'll never stop learning. This role gives you first-hand experience talking with leading researchers in the field, understanding their problems, and directly shaping the product direction.
• Our experienced founding team has successfully built and sold ML tools in the past at Figure Eight, and their deep knowledge of our industry, empathy for our users, and skillful management is driving W&B to success.
• Customers genuinely benefit from our tool. Here's a quote from Wojciech Zaremba, Cofounder and Robotics Lead, OpenAI: W&B allows to scale up insights from a single researcher to the entire team, and from a single machine to hundreds of them.