AI Engineer, Playable Ads - Tokyo Japan.
On-siteTokyo, Tokyo, Japan
Job Summary
Build and operate reliable, scalable ML and generative AI pipelines that power automated content creation, personalization, and optimization across creative formats. Productionize research prototypes by designing service contracts, containerized workers, and asynchronous orchestration while applying modern ML including LLMs and multimodal generation. Engineer workflows with schema validation, retries, and end-to-end tests, then define metrics for quality, latency, and cost using logs and evaluations. Collaborate with scientists, engineers, and designers to translate business needs into maintainable systems. This role sits at the intersection of applied ML and software engineering within Appier's Playable Ads team, which builds AI-powered ad experiences for global apps and games.
Required Qualifications
- Bachelor's or Master's degree in Computer Science, Machine Learning, Electrical Engineering, or a related field—or equivalent practical experience
- Three or more years of experience building production software or ML systems, with strong computer science fundamentals and a record of writing maintainable, tested code
- Strong Python skills and hands-on experience with PyTorch, TensorFlow, or equivalent ML tooling
- Practical experience taking models, prompts, or data workflows from prototype to production using APIs, containers, batch or stream processing, and cloud infrastructure
- Experience with cloud and ML platform tooling such as Docker, Kubernetes, GCP, CI/CD, workflow orchestrators, message queues, Spark, or OpenTelemetry
- Understanding of production ML concerns including reproducibility, data and model versioning, evaluation, monitoring, failure handling, latency, and cost
- Strong debugging and systems-thinking skills across distributed components such as queues, workers, storage, and external services
- Clear communication and cross-functional collaboration skills; comfortable using AI-assisted development tools responsibly while validating their output
Desired Qualifications
- Experience with generative media, creative optimization, advertising or MarTech, recommendation, or content-generation products
- Experience designing agent tools, feed-back-loop pipelines, rigorous evaluations, or online experiments—and translating research into measurable product impact
- Hands-on experience with generative AI or multimodal systems, such as LLMs, VLMs, image or video generation, RAG, tool use, or agent frameworks
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.