Machine Learning Engineer - Conversation AI

$140,100 - $210,100/Yr

Doordash - San Francisco, CA

Full-time - Mid Level

San Francisco, CA

Couriers and Messengers

About the position

As a Machine Learning Engineer at DoorDash, you will play a pivotal role in developing and enhancing the models that drive our three-sided marketplace, which includes consumers, merchants, and dashers. Your primary focus will be on building a leading industry-level Natural Language Processing (NLP) and Large Language Model (LLM) system that empowers communication across the DoorDash Journey. This position offers the opportunity to identify and prioritize machine learning investments within our conversation AI and personalization ecosystem, leveraging our robust data and infrastructure to create impactful models that serve millions of users. In this role, you will lead the development of DoorDash's support chatbot and LLM system, applying advanced techniques such as active learning, semi-supervised learning, weak label generation, and data augmentation strategies to enhance the support experience for consumers, dashers, and merchants. You will also drive the personalization of our issue prediction and resolution policies, utilizing recommendation and dynamic pricing modeling technologies to provide tailored solutions for our customers. Additionally, you will spearhead the creation of next-generation LLM AI Agent tools, including a Co-pilot system that transforms user interactions with our support system. Your responsibilities will include applying stratification, variance reduction, and advanced experiment design techniques to create A/B tests that efficiently measure the impact of your innovations while minimizing risks to the broader system. This role is not just about technical skills; it also involves strategic collaboration with an engineering lead and product manager to align machine learning initiatives with business metrics that drive growth.

Responsibilities

• Lead the development of DoorDash's support chatbot and LLM system.
• Apply LLM, active learning, semi-supervised learning, weak label generation, and data augmentation strategies to improve support experiences.
• Drive the personalization of DoorDash's issue prediction and resolution policies using recommendation and dynamic pricing modeling technologies.
• Spearhead the creation of next-generation LLM AI Agent tools, including a Co-pilot system.
• Apply stratification, variance reduction, and advanced experiment design techniques to create A/B tests to measure the impact of innovations.

Requirements

• 3+ years of industry experience developing optimization models with business impact.
• 1+ year(s) of industry experience serving in a tech lead role.
• M.S. or PhD. in Statistics, Computer Science, Math, Operations Research, Physics, Economics, or other quantitative field.
• Deep understanding of natural language processing techniques and procedures for acquiring and validating human-labeled data.
• Good experience in overall big data analysis and system integration with new ML solutions.
• Familiarity with programming languages such as Python and machine learning libraries like SciKit Learn and Spark MLLib.
• Experience in productionizing and A/B testing different machine learning models.
• Familiarity with advanced causal inference techniques and contextual bandit algorithms preferred.

Nice-to-haves

• Experience with advanced causal inference techniques.
• Familiarity with contextual bandit algorithms.

Benefits

• Comprehensive healthcare benefits.
• 401(k) plan with employer match.
• Short-term and long-term disability coverage.
• Basic life insurance.
• Wellbeing benefits.
• Paid time off and paid parental leave.
• Several paid holidays.
• Opportunities for equity grants.