Staff Machine Learning Engineer

This hybrid role requires working in the office two days per week.

With millions of diners, 70,000+ restaurant partners and 25+ years of experience, OpenTable, part of Booking Holdings, Inc. (NASDAQ: BKNG), is an industry leader with a passion for helping restaurants thrive. Our world-class technology empowers restaurants to focus on what matters most – their team, their guests, and their bottom line – while enabling diners to discover and book the perfect restaurant for every occasion.

Every employee at OpenTable has a tangible impact on what we do and how we do it. You’ll also be part of a global team and its portfolio of metasearch brands. Hospitality is all about taking care of others, and it defines our culture.

About the Role

The Data Science team at OpenTable supports a wide range of initiatives targeting diners, restaurants, and internal stakeholders. The team is expanding its capabilities across multiple areas, including building AI Agents to power restaurant search and discovery as well as AI-augmented products for restaurant management.

As a Staff Machine Learning Engineer , you will set the technical direction for how OpenTable builds, serves, and operates machine learning systems in production. You will partner with Machine Learning Scientists and engineers across the company to take models from experimentation to reliable, monitored production services --- and you will define the standards and patterns the rest of the team builds on.

This is a deliberately engineering-forward role. We are looking for an engineer who builds and operates production systems, rather than a modeller who deploys occasionally. The strongest candidates will bring hard-won judgment from more than one organization about how mature ML teams actually work, and the ability to apply it here. This posting is for an existing vacancy.

Key Initiatives

• AI Agents for restaurant discovery

• Personalized recommendations for diners

• Developing and serving high-throughput predictive models for strategic marketplace optimization initiatives

• Building and integrating tools into our agentic platform via LLM tool calls, MCP, and Agent-to-Agent protocols

• Multimodal understanding of restaurant content (text, images, geospatial)

• Creating an AI-powered platform for restaurant partners to gain insights into their business performance and diner demand

What You'll Own

• Technical direction for how models are served, deployed, and monitored; the architecture, the patterns, and the tradeoffs behind them.

• Production ML services that are high-throughput, low-latency, and observable, from design through operation.

• Engineering standards for ML systems: testing, CI/CD, observability, alerting, rollback, and on-call practice.

• Ambiguous, cross-team problems: scoping them with Product Managers and stakeholders, then ruthlessly prioritizing what the team actually builds.

Requirements

• 7+ years of professional software engineering experience, with a substantial portion spent building and operating machine learning systems in production.

• Breadth of industry perspective. You have seen how ML systems are built and operated at more than one organization, and can speak to industry-standard practices, common reference architectures, and where the real tradeoffs lie.

• Hands-on experience with a major cloud platform (AWS, GCP, or Azure) as a primary model-serving environment, including its managed services for deployment, scaling, and observability.

• Deep engineering fundamentals: distributed systems, service and API design, concurrency, latency and throughput tradeoffs, testing discipline, and genuine production ownership including on-call.

• Strong command of Python and proficiency in at least one strongly typed language (Java preferred).

• Demonstrated experience training, serving, and deploying ML models at production scale.

• Production MLOps ownership: model and feature monitoring, drift and data-quality detection, retraining and promotion workflows, versioning, safe rollout and rollback, and incident response when a model misbehaves.

• A track record of technical leadership: leading multi-quarter projects, influencing engineering decisions beyond your immediate team, and coordinating with Product Managers and other stakeholders.

Strong Preference

• Serving LLMs in production: inference infrastructure, GPU utilization, batching and caching strategies, quantization, and managing the latency/cost frontier (vLLM, TGI, TensorRT-LLM, or similar).

• Applied ML depth in ranking, recommendations, classification, NLP, RAG, and/or agentic systems.

• Kubernetes in production at meaningful scale.

• Experience developing ETL jobs (especially Spark) or data warehouse infrastructure.

• Familiarity with A/B testing design and analysis best practices.

• Experience introducing new tooling or platform capability to a team and driving adoption.

Tech Stack

We do not expect experience with everything on this list; it is here so you know what you would be working with.

• Pipelines: Spark, Airflow, EMR, SageMaker, Snowflake, S3, Delta Lake

• ML: PyTorch, XGBoost / CatBoost, LLMs, LangChain, LangSmith

• Deployment: Docker, Kubernetes, Helm, Prometheus, Graphite / Grafana

• Infrastructure: Kafka, ElasticSearch, Postgres, MongoDB, Redis, Qdrant

• Build: Poetry, FastAPI, Flask, Gunicorn / Uvicorn, Spring, Maven, TeamCity

Why This Team

The ML team at OpenTable has two opposing challenges which manifest themselves as opportunities:

• OpenTable is the world's leading provider of online restaurant reservations, seating more than 25 million diners per month via online bookings across approximately 70,000 restaurants. It has a massive wealth of diner and restaurant data going back over 25 years.

• OpenTable fields a lean team, with just over 1,600 employees globally. The ML team is currently fourteen people, but striving to grow.

As a member of the team, you will benefit from these factors because your projects will have sufficient data and usage to be interesting and have a meaningful impact, and you will have the opportunity to work on a variety of interesting projects across the company. However, you will be forced to think critically and ruthlessly prioritize, since the team has finite bandwidth. If this sounds like an interesting challenge, we look forward to hearing from you.

Benefits and Perks

• Work from (almost) anywhere for up to 20 days per year

• Focus on mental health and well-being:

• Company-paid therapy sessions through SpringHealth

• Company-paid subscription to Headspace

• Annual company-wide week off a year - the whole team fully recharges (and returns without a pile-up of work!)

• Paid parental leave

• Generous paid vacation + time off for your birthday

• Paid volunteer time

• Focus on your career growth:

• Development Dollars

• Leadership development

• Access to thousands of on-demand e-learnings

• Travel Discounts

• Employee Resource Groups

• 20 days of paid time off

• Private health and dental insurance

• Life and Disability insurance

The best connections happen face-to-face, whether you’re sitting down to dinner or having coffee with a coworker. That’s why OpenTable has adopted a hybrid workplace model. This role aligns with that approach, with an expectation of coming into the office two days a week—giving employees the best of both worlds: in-person collaboration and flexibility.

There are a variety of factors that go into determining a compensation range, including but not limited to external market benchmark data, geographic location, and years of experience sought/required.

…