At Cohere, we strive to continually improve our large language models. Academic research and real-world experience has demonstrated that high quality, diverse datasets can contribute as much to the performance and capabilities of LLMs as the underlying model architecture and training regimen. We at Cohere believe data will play a central role in accelerating the advancement of our already world-class language models.

Data is therefore critical to our success. Our ability to acquire data that is accurate, relevant, and timely is key to our ability to improve the quality of our models. We strive to continuously improve our data acquisition processes and systems to ensure that we have the data we need to stay competitive and meet the needs of our customers. We run frequent experiments to learn more about the role of data for model quality, from data mixtures, to cleaning techniques, to quality control.

This role will be part of the Data Acquisition team, which broadly provides data for training models and is responsible for building and maintaining the infrastructure that acquires, cleans, and formats data for model training. We are looking for a technically skilled, resourceful problem-solver who is able to work in areas of ambiguity and find efficient and sometimes creative solutions. The main responsibility of this role is to improve our internal data acquisition infrastructure, which includes data crawlers, formatters, and integrations with data providers. This role would also work closely with different teams at Cohere to support their data acquisition needs, as well as engage in more experimental work to develop highly informative data signals.

Please Note: We have offices in Toronto, Palo Alto, and London but embrace being remote-first! There are no restrictions on where you can be located for this role.

As a Senior Software Engineer specializing in Data Acquisition, you will:

• Acquire, prepare, and integrate high-quality datasets into model training and evaluation
• Design, build, and optimize large-scale data pipelines for ingesting and processing data from diverse sources
• Collaborate with research and product teams to identify, prioritize, and secure new data sources
• Develop tooling and infrastructure to systematically evaluate and distill large datasets
• Mentor junior team members and contribute to their professional growth

You may be a good fit if:

• You have 5+ years of experience in software development, data engineering, or machine learning
• You have written, debugged, and optimized Spark code over petabyte-scale datasets
• You have proficiency in Python
• You have successfully implemented data quality solutions in production
• You are familiar with orchestration engines like Airflow, Dagster, and Prefect
• You have trained classifiers to perform large-scale data labelling
• You are comfortable using embeddings or language models as part of your workflow
• You have strong communication and problem-solving skills, and prefer using the right tool for the job even when it’s outside your wheelhouse
• You have a demonstrated passion for applied NLP models and products