Our team’s mission is to empower Mercari’s marketplace with machine learning systems that improve discovery, trust, and ease of use. We build models that drive personalization, ranking, and fraud prevention to enhance the buying experience, while also developing ML systems that simplify the seller journey through solutions such as automated categorization and pricing support. By combining scalable infrastructure, cutting-edge ML techniques, and deep market understanding, we aim to create a seamless and reliable experience for millions of buyers and sellers.
The Machine Learning team is responsible for the end-to-end development and operation of ML pipelines, from data collection and feature engineering to model training, deployment, and monitoring. We ensure our models not only deliver measurable impact on user engagement and conversion, but also operate at scale with stability, fairness, and efficiency.
Recent or ongoing initiatives include:
- Building scalable ML pipelines for ranking, recommendation, and personalization.
- Deploying real-time inference systems that improve user trust and reduce fraud.
- Leveraging large language models (LLMs) and generative AI to enhance search recall, content understanding, and overall user experience.
- Optimizing experimentation frameworks to accelerate product iteration and innovation.
- Collaborating with search, backend, and frontend teams across the US and Japan to deliver high-impact ML features at global scale.
Unique Challenges
- Machine learning is core to Mercari’s marketplace, powering search ranking, personalized listing recommendations, and fraud detection which directly impact user trust and growth. You will own the continuous improvement cycle of ML systems, from ideation to production, with measurable business outcomes.
- Collaborate closely with product managers, data engineers, backend engineers, and QA engineers to design advanced ML solutions that balance accuracy, latency, and scalability.
- Architect and operate highly scalable ML services and pipelines to support rapid user and product growth in the US market.
- Proactively contribute to Mercari’s engineering culture by sharing knowledge, proposing improvements, and mentoring peers.
- Develop a deep understanding of user behavior and market trends in the US to ensure ML models align with customer expectations and business goals.
- Work in a globally distributed team, collaborating with teammates in both the US and Japan, navigating cultural and time zone differences to deliver high-impact results.
Responsibilities
- Lead the end-to-end ML model lifecycle: independently identify machine learning opportunities through data and metric analysis, define and track KPIs, and ship iterations that align with product and business goals.
- Design, develop, and maintain ML pipelines, including feature engineering, model training, deployment, and monitoring.
- Implement scalable inference services and APIs for real-time and batch predictions.
- Improve model accuracy, inference speed, and robustness through experimentation, hyperparameter tuning, and feature optimization.
- Ensure reliability through comprehensive automated testing, observability, and reproducibility of ML experiments.
- Mentor junior engineers, lead code and model reviews, and actively contribute to architectural decisions and technical documentation.
- Collaborate with cross-functional teams across product, engineering, and operations to deliver impactful ML solutions.
Requirements
- Strong hands-on experience across the machine learning model life cycle: training, deployment, monitoring, and optimization.
- Practical experience leveraging computer vision and natural language processing techniques in production ML systems.
- Ability to independently analyze data and model metrics to ship measurable improvements in production systems.
- Bachelor’s degree in Computer Science, Data Science, Mathematics, or a related field (or equivalent practical experience).
- 5+ years of professional experience developing and operating large-scale ML pipelines and/or backend services in high-traffic production environments, including optimizing models for latency, scalability, and cost efficiency.
- Experience with large language models (LLMs) and generative AI, including techniques such as prompt engineering, fine-tuning, vector search integration (RAG), and responsible production deployment.
- Experience with ML frameworks and pipelines (TensorFlow, PyTorch, scikit-learn, MLflow, Kubeflow, or similar).
- Strong programming expertise in Python; familiarity with Go or PHP is a plus.
- Excellent English communication skills, with the ability to collaborate effectively across functions and regions.
- Demonstrated ability to mentor and guide junior engineers.
- Business level English ability (CEFR B2 or higher) required
Nice to haves
While not specifically required, tell us if you have any of the following.
- Experience deploying and scaling ML services in production environments, including cloud platforms (GCP, AWS, or Azure), containerization (Docker, Kubernetes), CI/CD, Infrastructure as Code (Terraform), and observability (Prometheus, Grafana).
- Familiarity with data engineering practices, including feature stores, data preprocessing, ETL pipelines, large-scale data management, and real-time streaming or event-driven architectures (e.g., Kafka, Pub/Sub).
- Domain knowledge of marketplace or e‑commerce platforms.
- Contributions to open-source projects in ML or related areas; or public technical engagement through blogs, talks, or conferences.
- Experience working within large, cross-functional, and geographically distributed teams.
- Japanese ability
