Machine Learning Engineer (Gradescope)

At Turnitin, our mission is to ensure academic integrity, help students get informative feedback and surface insights about teaching and learning. Our products are used at over 15,000 schools in over 150 countries, covering over 30M students. The company was founded almost 20 years ago, and was recently acquired by Advance (owner of Conde Nast and Discovery Network). Turnitin is headquartered in Oakland, with offices in Pittsburgh, Newcastle (UK), Utrecht (Netherlands), Seoul (South Korea), and Melbourne (Australia).

Machine Learning is integral to the continued success of our company. Our product roadmap is exciting and ambitious. You will join a team of curious, helpful, and independent scientists and engineers, united by a commitment to deliver cutting-edge, well-engineered ML systems.

We are in a unique position to deliver Machine Learning useful to hundreds of thousands of instructors teaching millions of students around the world. Over 1B papers have been submitted to the Turnitin platform, and over 100M answers have been graded on Gradescope by Turnitin. Machine Learning powers automated feedback on student writing, investigates authorship of student writing, groups handwritten student answers by content, and plays a crucial role in many back-end processes.

We expect Machine Learning Engineers to be strong software engineers with a passion for machine learning methods and applications. You will focus on ensuring turn-key model training and bringing trained models to production at scale, with significant capacity for dataset construction, and building web interfaces for data exploration and system monitoring.

Day-to-day, your responsibilities are to:

• Write and review clean, efficient, and modular code, with automated tests and appropriate documentation.
• Answer questions and make trainable datasets from raw data, using efficient SQL queries and scripting languages, visualizing when necessary.
• Process data efficiently, building distributed systems if necessary.
• Achieve reproducible model training distributed over multiple GPUs, CPUs, and machines.
• Deliver prediction-serving system with required scale, uptime, and monitoring.
• Stay up to date with technology, make good technological choices, and be able to explain them to the organization.

Required Qualifications:

• Experience with the below responsibilities.
• At least 3 years of relevant development experience.
• Strong software engineering fundamentals (we use Python, Unix-based systems, git, and github for collaboration and review).
• Essential dev-ops skills (we use on-prem hardware, AWS EC2/Batch/Lambda).
• Essential machine learning development skills (we use sklearn, tensorflow, pytorch, jupyter).

Desired Qualifications:

• Degree in Computer Science, Statistics, Applied Mathematics, or a related field.
• Experience or interest in education and machine learning methods for education technology.
• Experience with cloud-based workflows.
• Familiarity in front-end frameworks (e.g. React, Javascript) to build interfaces as needed.

Characteristics for success:

• Enthusiasm for solving challenging problems and not being thwarted by obstacles.
• Ability to work independently with a consistently high output.
• Excellent written and verbal communication skills (Upper-intermediate and Fluent English).
• Strong curiosity about the problems at hand, the field at large, and the best solutions.
• Strong system-level problem solving skills.

We offer:

• Professional growth opportunity.
• Generous holiday allowance.
• Free English classes.
• Young and friendly team.
• Close-knit family atmosphere.
• Flexible schedule.
• Comfortable and modern office.