Staff Machine Learning Engineer

Zendesk builds software for better customer relationships, empowering organizations to improve customer engagement and understand their customers. Zendesk products are used by more than 100,000 paid customer accounts in over 150 countries and territories. The company is based in San Francisco with operations in the United States, Europe, Asia, Australia, and South America. The team works on routing and presence products, focusing on evolving from a rules-based engine to an agentic routing engine with intelligent, adaptive decisioning.

• Apply classical ML including probabilistic models, feature engineering, optimization, and model selection with understanding of why approaches work.
• Work hands-on with generative AI and LLMs including fine-tuning, RAG, prompt engineering, and agentic patterns.
• Design and run experiments end to end with power analysis and guardrail metrics.
• Ship ML in production considering latency, throughput, drift detection, retraining triggers, versioning, rollback, and operational realities.
• Estimate ML project effort and make product tradeoffs considering timeline, scope, quality, and cost.
• Communicate designs and decisions clearly to engineers and scientists.
• Motivate research outcomes in terms of product strategy and long-term business goals.
• Collaborate deeply with backend engineers, contributing to services written in Java and Scala.
• Explain model behavior, uncertainty, and limitations to non-technical stakeholders.
• Balance focused technical delivery with supporting teammates' ML questions.
• Identify and decompose requirements using scientific method and group problem-solving.
• Engage regularly in self-learning of ML, statistics, software engineering, and relevant research.

• Own the models and algorithms at the heart of Routing & Presence, ensuring quality, reliability, and interoperability.
• Plan and scope ML work with observability and iterability designed from the start.
• Propose and evaluate alternative design approaches with tradeoffs and reasoning.
• Design experimentation frameworks (offline evaluations and online A/B tests) tailored to routing outcomes.
• Lead innovation sessions with Product to shape what is built.
• Identify new data sources needed for learning and evaluation.
• Combine multiple signals into dynamic routing decisions with explainability and guardrails.
• Shape the boundary between classical ML, LLM-driven components, and rules.
• Anticipate issues across the full lifecycle and build mitigations.
• Collaborate across multiple teams to define problems and set high-level solution directions.
• Mentor scientists and engineers to raise ML literacy and support teammates.
• Plan for long-term scalability and maturity of the agentic routing engine.
• Share best practices with the wider science and engineering community at Zendesk and beyond.

Zendesk is an equal opportunity employer committed to diversity, equity, and inclusion. Employment consideration is without regard to race, color, religion, national origin, age, sex, gender identity or expression, sexual orientation, marital status, medical condition, ancestry, disability, military or veteran status, or any other protected characteristic. Reasonable accommodations are available for applicants with disabilities.