Product Engineer (AI/ML), CS & TnS ML - Mercari

Employment: Full-time Experience: Mid-Senior level 本ポジションは日本語JDの用意がありません。 Product Engineer (AI/ML), CS & TnS ML - Mercari Employment Status:　Full-time Work Hours: Full Flextime (no core time) 　 Office: Roppongi For more details, see the Overview of Our Positions section on our Careers site. 　 About Mercari Circulate all forms of value to unleash the potential in all people "What can I do to help society thrive with the finite resources we have?" The Mercari marketplace app was born in 2013 out of this thought by our founder Shintaro Yamada as he traveled the world. We believe that by circulating all forms of value, not just physical things and money, we can create opportunities for anyone to realize their dreams and contribute to society and the people around them. Mercari aims to use technology to connect people all over the world and create a world where anyone can unleash their potential. For more information about Mercari Group’s mission, see Mercari’s Culture Doc 　 Organization/Team Mission Mercari Engineering Principles Mercari Engineering Principles are a shared understanding that serves as the foundation of engineering beliefs and behavior at Mercari. The Engineering Principles are designed to complement the organizational identity (Mercari’s mission, values, and culture) from an engineering viewpoint. These principles ultimately help us achieve Mercari’s mission by defining the ideal state we seek to realize in the long term. Passion For The Product Grow Together Solve Through Mechanisms Collaborate Openly For more details, please see the following link: Engineering Culture CS & TnS ML team The Customer Support (CS) & Trust and Safety (TnS) Machine Learning (ML) Team keeps Mercari safe for tens of millions of users. We build machine learning models that keep the marketplace safe and reduce manual operations, by detecting prohibited items, harmful behavior, fraud and abuse, and by automating replies to customers. Our models run in real time and decide, within milliseconds, what is safe to show or complete. We own the full life of each model: from the first specification, through training and evaluation, to serving, monitoring and improvement in production. See here for more information about our mission and values. 　 Work Responsibilities End-to-End Ownership: Own an ML problem from specification to delivery. Work with product and operation teams to turn a rule or a risk into a measurable target. Then build, ship and maintain the model that meets it. Model Development: Design, train and evaluate models on image, text and behavioral data (multimodal deep learning, Japanese NLP, tabular and sequence models). Apply them across our domains: item moderation, comment moderation, transaction message moderation, customer auto-reply, and payment fraud prevention. Production Serving: Deploy models as reliable, low-latency real-time services (for example ONNX, TensorRT, gRPC on Kubernetes). Build the CI/CD, rollout and rollback paths. Monitor model health, latency and cost. Evaluation and Experimentation: Build offline evaluation that matches production conditions. Verify changes with shadow runs and A/B tests. Detect drift and quality regressions early, with dashboards and alerts. Data and Labels: Improve the data behind each model. Work with moderators, customer reports and labeling tools. Find and handle label noise, bias and gaps. Initiative and Proposals: Find where ML can reduce risk or manual work, propose the initiative with a clear business case, and drive it to delivery. Rapid Response: Respond quickly to urgent requests, such as a viral incident or a policy change. Deliver a safe first fix fast, then improve it. Working Across Boundaries: Do what the problem needs, not only what your title says. This can include data pipelines, backend services and infrastructure. Use AI coding tools to move faster, and stay responsible for the quality of the result. 　 Unique Challenges Rare and Noisy Positives: Harmful content is a tiny share of everything on the marketplace. You will build models and evaluations that stay trustworthy under extreme class imbalance and label noise. Fast-Moving Threats and Urgent Requests: Abusers change tactics, a post can go viral, and policies change. You will adapt models and thresholds quickly, without hurting precision for the many legitimate users. Real-Time at Scale: Our decisions must come within strict latency and cost limits for tens of millions of users. You will choose and tune methods that are accurate and also fast enough to serve. Many Domains, One Platform: Items, comments, messages, payments and customer support each have their own data and risks. You will share methods, tooling and serving across them, and learn new domains as the need appears, with support from senior teammates. 　 Qualifications Required Experience/Skills Shared belief in the mission and values of Mercari Group. Proficiency in Python. Daily use of AI coding tools in your development work, with the habit of reviewing and verifying what they produce. Solid foundation in machine learning and deep learning, with hands-on experience training, evaluating and debugging models (work, research or personal projects). Eagerness to learn quickly, take end-to-end ownership, and work beyond the boundary of your own area of expertise. Strong written and verbal communication, with the ability to propose ideas and align stakeholders (product, operations, policy). Preferred Experience/Skills 2+ years of professional experience in machine learning or software engineering, with models in production. Experience in one or more of: computer vision, NLP (Japanese is a plus), fraud or anomaly detection on tabular and sequence data (for example gradient boosting, feature engineering), agentic AI. Experience serving models in production: Docker, Kubernetes, gRPC or REST, model optimization for latency (for example ONNX, TensorRT, quantization). Experience with cloud and data tools: GCP, SQL and BigQuery, and ML pipelines (for example Kubeflow, Vertex AI, Apache Beam). Experience designing offline evaluations and A/B tests, and monitoring models in production. Knowledge of trust and safety, content moderation or fraud prevention. Master's or Ph.D. in a related technical field, or the habit of following new research and applying it. Language English: Proficient (CEFR B2). Required. Japanese: Basic (CEFR A2). Required. For details about CEFR, see here . 　 Learn More About Mercari Group Careers site: https://careers.mercari.com/en/ Mercan: https://mercan.mercari.com/en/ Social media: X / Linkedin 　 Recruiting at Mercari At Mercari Group, we value empathizing with and embodying the mission and values ​​of the Group and each company. To promote the creation of an organization that maximizes the total amount of value exhibited by all members, we would like to understand the experience and skills of each candidate as accurately as possible. Recruiting cycle at Mercari Group Application screening Skill assessment: For engineering positions, you will be asked to complete a skill assessment on HackerRank or GitHub. For non-engineering positions, you may be asked to complete an assessment depending on the position. (The timing of the assessment may coincide with the interview process.) Interview: The number of interviews may vary depending on the position. Reference check: We will ask for online references around the timing of the final interview. Offer: Offers will be determined carefully in consideration of the final interview and the reference check. 　Learn more about our recruiting process here . 　 Equal Opportunity Hiring Here at Mercari, we work to realize a world in which no one’s potential is limited by their background and everyone has the opportunity to freely create value. We also firmly believe that a mindset of Inclusion & Diversity is essential for us to achieve our mission. This, of course, extends to our hiring practices as well. Mercari is committed to eliminating discrimination based on age, gender, sexual orientation, race, religion, physical disability, and other such factors so that anyone who shares our mission and values can join us, regardless of their background. For more details, please read our I&D statement . Please read and acknowledge our Privacy Policy prior to submitting your application.