Staff Machine Learning Engineer

Grailed is looking for a Staff Machine Learning Engineer to help us build the models and systems that connect buyers to the inventory they're looking for — and surface things they didn't know they wanted. Our data sits at the center of a complex peer-to-peer marketplace, and the ML layer is what turns a decade of behavioral signals into better search, smarter recommendations, and a marketplace that gets sharper over time.

This is a hands-on technical role for an engineer who takes end-to-end ownership seriously — from architecture through production operation — and who is energized by working on a small, focused team where the infrastructure is still being built and the decisions made now have lasting consequences.

The strongest candidates will bring production instincts alongside technical depth: the kind of engineer who isn't done when the model trains, and who treats monitoring, retraining, and reliability as part of the job, not a follow-on task.

• Own the full lifecycle of predictive models in production — architecture, training pipelines, inference infrastructure, deployment, and ongoing model health
• Build and operate the systems that route model outputs into live product surfaces: search ranking, recommendations, feed ordering, and related user-facing experiences
• Establish and maintain model monitoring, alerting, drift detection, and retraining cadences — the feedback loops that keep deployed models accurate over time
• Partner closely with Data Science, Data Engineering, Product Management, and backend engineering to move work from validated approach to production system
• Own the decision-making process on whether to leverage ML infrastructure & expertise from our parent company, GOAT Group, and when to advocate for building in-house solutions.
• Contribute to ML infrastructure decisions — serving architecture, feature computation, pipeline orchestration — with an eye toward what scales as the team and model count grows
• Set technical standards and raise the bar for how ML systems are built, evaluated, and operated across the pod

• 7+ years of engineering experience, with substantial depth in production machine learning systems.
• Demonstrated end-to-end ownership: training pipelines through deployed inference, not just modeling.
• Advanced knowledge of ML, AI and statistical models, as well their application in e-commerce settings.  
• Strong proficiency in Python; SQL; DBT; airflow or similar.
• Solid software engineering fundamentals.
• Experience with ranking, retrieval, or recommendation systems.
• Demonstrated expertise with ML lifecycle tooling — experiment tracking, model versioning, pipeline orchestration, drift detection — and comfort working with modern data infrastructure (cloud warehouse, search/retrieval systems).

• Takes ownership of developing repeatable end-to-end processes, not just outcomes
• Evaluates technical approaches against production constraints — latency, reliability, retraining cost — not just offline metrics
• Brings judgment to architecture decisions: knows when to reach for a complex approach and when a simpler one is the right call
• Treats model health as a permanent responsibility, not a launch milestone
• Communicates clearly with non-technical partners — can translate model behavior, tradeoffs, and timelines into terms that product and business stakeholders can act on
• A willing collaborator who keeps people informed and works through ambiguity without going quiet
• Genuine curiosity about the domain — fashion, resale, taste — and the specific ML problems it creates

• Experience with semantic enrichment, NLP, or multi-modal ML in a production context
• Genuine curiosity about the domain — fashion, resale, style — and the specific ML problems it creates

One last thing — add a quick note at the bottom of your resume (1–3 lines): what drew you to Grailed and this role, and a recent buying or selling experience on any marketplace and what made it stand out or fall flat. There are no wrong answers — we actually read all of these.