Description
Stripe is a financial infrastructure platform for businesses. Millions of companies,from the world’s largest enterprises to the most ambitious startups,use Stripe to accept payments, grow their revenue, and accelerate new business opportunities. Our mission is to increase the GDP of the internet, and we have a staggering amount of work ahead.
The Applied ML team at Stripe aims to reform how users interact with Stripe by automating easy tasks and assisting with difficult ones. Examples include helping users resolve issues with Stripe faster or making it easier for users to sign up and navigate Stripe. The team uses the latest Large Language Models (LLMs) and fine-tunes its own models.
As a Machine Learning Engineer, you will analyze opportunities, propose ideas, train and evaluate ML models, run experiments, and deploy everything to production. You will also contribute to and influence ML architecture at Stripe and be part of a larger ML community.
Responsibilities:
- Develop pipelines and automated processes to train and evaluate models in offline and online environments
- Integrate ML models into production systems and ensure their scalability and reliability
- Collaborate with product and strategy partners to propose, prioritize, and implement new product features
- Engage with the latest developments in ML/AI and transform innovative ML ideas into productionized solutions
The team operates fluidly, tackling problems such as:
- Evaluating systems offline and online
- Improving performance to match and beat human standards
- Ensuring model quality doesn’t degrade online
- Fine-tuning LLMs for better performance
- Investing in the right OSS and in-house platforms
Requirements:
- At least 3 years of experience shipping ML systems in production
- High standards for working with production systems
- Ownership and drive to take projects to business impact
- Collaborative environment
Preferred qualifications:
- 5+ years of experience in full-time software development roles
- Experience shipping LLM integrations to user products with high quality
- Operating in highly ambiguous environments
- Driving hypotheses from data