2 months ago
Amsterdam, NetherlandsSenior
Responsibilities
- Design and architect robust multi-step, multi-agent systems across customer-facing eBay surfaces.
- Build systems in which agents coordinate, delegate, communicate, critique reasoning, and solve complex user intents.
- Research and implement agent interaction patterns such as structured communication protocols, reflection loops, evaluator-generator frameworks, and tool-augmented reasoning.
- Optimize latency and cost while addressing safety, guardrails, and cascading failures in production multi-agent systems.
- Collaborate with applied researchers, engineers, product, design, analytics, and globally distributed Search & Recommendations teams.
- Use large-scale unstructured and multimodal inventory data and A/B testing to improve marketplace metrics and autonomous user experiences.
Requirements
- Master’s or PhD in Computer Science or a related area.
- At least 5 years of relevant work experience in Machine Learning, AI, or ML Engineering.
- Demonstrated experience researching, building, or studying systems where multiple agents interact, exchange information, critique reasoning, and coordinate decisions over time.
- Extensive development experience, preferably in an ML/AI technology environment, including Python and PyTorch.
- Experience using cloud services, big data pipelines, and databases in an industrial setting.
- Experience in Natural Language Processing and industrial recommender systems.
Categories
About TCGplayer
TCGplayer operates an online marketplace and software tools for buying and selling trading card games and related collectibles, serving hobby shops and individual sellers. It generates revenue through marketplace fees, seller subscriptions, and fulfillment programs such as TCGplayer Direct. An eBay subsidiary, the company provides authentication and logistics services that help stores list inventory at scale and reach buyers across the U.S. and internationally.