Staff Machine Learning Engineer

Description

The company is hiring a Staff MLE for its Discovery team to own recommendations and personalization across the consumer site: the homepage feed, product recommendations, search, and the ML systems that help families build registries.

The company was built on editorial picks chosen by people with deep baby gear knowledge, and millions of families trust that. It is now adding ML personalization on top of that foundation, using one of the richest first-party datasets in parenting.

The work is early. A small Discovery team has first retrieval and reranking models live on parts of the site and runs A/B tests regularly. The company wants someone who has seen personalization work well at scale and can set the technical direction from here.

Registry building is central to the product. Parents build lists of dozens of items, from swaddles to strollers, and a friend or relative will buy them and a baby will use them. Because the registry is universal, it often pulls from many retailers. The problem involves latent intent, life-stage changes, several people shaping one gift, millions of completed registries as explicit signal, and cross-retailer data.

This role is not a fit for someone who wants to tune models at the margins of a mature ML organization. It suits someone who has worked on a strong recommendations team and wants to build from zero to one at a younger company.

Responsibilities

• Serve as technical lead for personalization on the Discovery team, working with a PM, an engineering manager, a senior MLE and full-stack engineers.
• Set the direction for the models over the next one to two years, decide the order of bets, and keep building hands-on.
• Own product recommendations on every surface: product detail pages, add-next suggestions after an item goes into a registry, aesthetic or vibe representations of products, and the shared models beneath them.
• Decide which products, editorial pieces and social content each family sees, and in what order.
• Bring what the company knows about a family into search results, and help decide how far search should share retrieval and ranking with recommendations.
• Work with the team behind the AI registry assistant so its product suggestions use the same models and signals as the rest of the site.
• Take an unclear business problem from first sketch to production model, and own whether it moved the metric.
• Make the modeling and architecture decisions that span surfaces and are costly to reverse.
• Design offline and online evaluation that shows whether a model is good before and after launch.
• Judge when a rule or simple baseline is the right first step and when to move to more advanced methods.
• Work with product, design and data to shape what is worth building.
• Coach engineers on the team and act as the ML expert and resource for the wider engineering organization.

Requirements

• You have shipped recommendation or personalization systems that reached real users at scale in a consumer product, and you can show their business impact.
• You did that work on a team that did ML well, so you know what good looks like and want to bring it to a company earlier in its ML journey.
• You know recommender fundamentals, such as candidate generation versus ranking, offline versus online evaluation, cold-start handling, and correcting for position and popularity bias.
• You are fluent in the Python ML stack and can own models through deployment and monitoring, including contributing to the application backend.
• You can take an ambiguous problem and start making progress before someone gives you the full picture.
• You understand the tradeoff between effort and effectiveness in early-stage ML, and you can argue for a rule over a model, or for a model that is good enough to learn from.
• You focus on outcomes and users. You measure yourself by business impact and can tie a model change to user experience, registry adds, conversion or revenue.
• You already use AI daily to work faster and improve your output, and you keep up with what is coming next.

Conditions

Pay ranges are published. For a US-based Staff Engineer, the starting base salary is $233,500 to $290,700, with a target annual bonus of 20 percent of base, for total target cash of roughly $280,200 to $348,840.

Meaningful equity is included. Where you start in the range depends on your experience, and pay rises with performance and scope.

AI is the default way of working. Engineers run agentic sessions for most tasks and build the supporting pieces, including eval harnesses, curated context, custom review skills and fast CI. Agents triage incidents and handle a large share of support, and a person always owns the result.

The architecture is deliberately simple: one Rails monolith, MySQL, and few moving parts, which keeps the system easy for both people and AI to reason about.

The stack includes Rails, Packwerk, React, TypeScript and Sidekiq; Swift for iOS and Kotlin for Android; MySQL, Snowflake, dbt, Airflow, Weaviate, WarpStream, Hex and Sigma; deep learning, matrix factorization, retrieval and ranking, AWS SageMaker and MLflow; AI tools including Claude Code, Devin, Linear, CodeRabbit, Warden, LangChain, Bedrock AgentCore and Maxim; AWS, Kubernetes, Terraform, GitHub Actions, Datadog, incident.io and LaunchDarkly; and integrations with Shopify for payments and Iterable for CRM.

The company was founded when an engineer could not find the registry she wanted, and engineers still build for families. About 10 million people give gifts through the company each year.

The company reported over $750M in 2025 revenue, up 45 percent from the prior year. It has been profitable for eight years and remains independent. The engineering team is about 65 people.

The role is remote-first across the US and Canada, and the whole company meets in person twice a year. Engineering pods have three to five people. You will work directly with product, design, data and business partners, and you will regularly talk to customers through interviews, session recordings and ride-alongs with support.

Interviews are recorded and transcribed to evaluate candidates, in line with applicable privacy laws. You may use AI in the process, but your submissions and what you say should be your own thinking. If you have a family member or close relationship with a current employee, tell your recruiter.

Benefits

• Company-paid medical coverage, with dental and vision fully covered
• 401(k) with a company match
• Generous paid parental leave for birthing and non-birthing parents, plus a gradual return-to-work program
• Winter Wonder Week, a paid company-wide week off at the end of the year
• Remote-work stipend
• Mental-health and wellness support