(Senior) Machine Learning Scientist (LLM & Agents)
Salary not listed • Taipei, Taiwan • Full-time
Posted 1y ago
This job has been open for more than 60 days. It may no longer be actively hiring, so check with the company before putting a lot into an application.
About the job
Appier is seeking a Machine Learning Scientist specializing in LLMs and Agents to join the Enterprise Solution Science Team. This role involves applying advanced ML technologies to marketing challenges, leveraging omnichannel customer data. You will be responsible for the entire lifecycle of agent features, from initial concept and design through to production deployment, quality measurement, and ongoing improvement. The team focuses on developing AI agents that can automate marketing tasks, generate business strategies, answer complex data-driven questions, and enhance sales and customer service interactions.
Responsibilities
- Design, develop, and productionize enterprise AI agents for marketing, strategy, data analysis, and sales/service.
- Build agent capabilities as reusable skills and MCP tools, designing efficient tool interfaces.
- Integrate predictive models into agent workflows for next-best-action suggestions.
- Develop evaluation frameworks (test sets, LLM-as-a-judge, production trace review) to ensure quality.
- Optimize agent performance for speed and cost reduction without compromising answer quality.
- Write design documents, create reviewable PRs, and conduct code reviews.
- Collaborate with Product Managers, engineers, and QA throughout the development lifecycle.
Requirements
- Bachelor's degree or higher in Computer Science, Machine Learning, Statistics, or a related field.
- 2+ years of experience in ML or software engineering.
- Experience building at least one LLM or agent application end-to-end, including evaluation and improvement.
- Hands-on experience with MCP and agent skills.
- Experience with at least one agent harness (e.g., Claude Agent SDK, OpenAI Codex, Google ADK) and understanding of its impact on agent behavior.
- Ability to design evaluations for non-deterministic LLM output.
- Strong SQL skills and data reasoning capabilities.
- Proficiency in profiling and reducing agent latency and cost.
- Experience writing production-quality Python code with tests.
- Familiarity with AI-native development and coding agents.
- Clear written communication skills for design docs and PR descriptions.