DoorDash is looking to develop comprehensive AI/ML services to create highly accurate tax categorization systems for billions of restaurant and non-restaurant items processed through their platform.
Requirements
- Expertise in applied machine learning, particularly in deep learning, natural language processing, large language models (LLMs), and multi-modality models.
- Proficiency in Python, with experience using Spark, PyTorch, or TensorFlow preferred.
- Familiarity with Kotlin/Scala is advantageous.
Responsibilities
- Lead the development of innovative ML-driven tax catalog solutions, utilizing Generative AI to manage and organize tax categorization information efficiently.
- Design and implement ML products to address large-scale categorization challenges at the transaction level for DoorDash.
- Collaborate with engineering and product teams to shape the product roadmap, leveraging AI/ML capabilities to meet business objectives.
- Own the entire modeling lifecycle, including feature engineering, model development, deployment, experimentation, monitoring, explainability, and maintenance.
- Identify and explore new opportunities where AI/ML can be used as a strategic lever to expand into new markets, regions, and business lines.
- Stay abreast of emerging AI/ML trends and technologies, integrating them into existing systems to enhance performance and capabilities.
Other
- 6+ years of industry experience leading and developing advanced machine learning models with measurable business impact, including deploying solutions to production environments.
- Master’s or PhD degree in Statistics, Computer Science, Economics, Mathematics, Physics, or other quantitative fields.
- Strong ability to communicate complex technical concepts to non-technical stakeholders effectively.
- Demonstrated ability to take ideas from conception through rigorous testing to impactful results, with a focus on continuous improvement.
- Growth-minded, collaborative attitude with a passion for leveraging data and AI to create meaningful impact.
