Appier(Senior) Machine Learning Scientist (LLM & Agents)
Appier is hiring a (Senior) Machine Learning Scientist (LLM & Agents) in Taipei, Taiwan. Level rates it ; you can apply on Appier.
Level's read on this job
- On Level's scale, this role is Builds AI ●●●●: the job is building AI systems.
- Tools and skills the posting cites include OpenAI, Claude and Databricks.
- Level classifies it as a senior role.
- Level lists 66 other open roles at Appier.
AI in this role
About Appier
Appier is a software-as-a-service (SaaS) company that uses artificial intelligence (AI) to power business decision-making. Founded in 2012 with a vision of democratizing AI, Appier’s mission is turning AI into ROI by making software intelligent. Appier now has 17 offices across APAC, Europe and U.S., and is listed on the Tokyo Stock Exchange (Ticker number: 4180). Visit www.appier.com for more information.
About the Role
We are looking for a Machine Learning Scientist (LLM & Agents) to join the Enterprise Solution Science Team. This team focuses on applying cutting-edge ML technologies to real-world marketing problems by combining them with omnichannel customer data. In this role, you will own agent features end to end — from problem definition and design, to shipping, to measuring and improving quality in production.
What You’ll Work On
- You will contribute to the design, development, and productionization of one or more enterprise AI agents, including:
- Marketing Agent: Empower marketers to improve conversion rates and drive revenue growth.
- Strategy Agent: Transform raw data into actionable, revenue-driving strategies and high-potential audience segments.
- Data Analysis Agent: Answer business questions by planning multi-step analyses and running SQL over large customer datasets, with every number traceable to the query that produced it.
- Sales and Service Agent: Drive sales conversion, increase average order value (AOV), and deepen brand engagement and customer loyalty through real-time, 24/7 AI-powered support.
- Build agent capabilities as reusable skills and MCP tools, and design tool interfaces that agents use correctly and efficiently
- Bring predictive models (e.g. propensity, churn, recommendation) into agent workflows and next-best-action suggestions
- Build evaluation — test sets, LLM-as-a-judge, production trace review — and use it to decide what ships
- Make agents faster and cheaper without losing answer quality, through context management, caching and model choice
- Write design docs, break work into reviewable PRs, and review teammates’ code
- Work closely with PMs, engineers and QA from planning to release
What We’re Looking For (Minimum Qualifications)
- Bachelor’s degree or above in Computer Science, Machine Learning, Statistics, or a related field
- 2+ years of experience in ML or software engineering roles
- Has built at least one LLM or agent application end to end — at work, in research, or as a side project — and can explain how its quality was evaluated and improved
- Hands-on experience building with MCP and agent skills (tools, instructions and reference files an agent loads on demand)
- Hands-on experience with at least one agent harness (the runtime that drives the agent loop of model calls, tools and context), e.g. Claude Agent SDK, OpenAI Codex, or Google ADK — and an understanding of how the harness, tool design and context shape agent behavior
- Can design evaluation for non-deterministic LLM output, including test sets, metrics, and the limits of LLM-as-a-judge
- Strong SQL and data reasoning: can write and review analytical queries and tell when a result is wrong
- Can profile and reduce end-to-end latency and cost of an agent — e.g. fewer model round-trips, better tool design, caching, model choice — without hurting answer quality
- Writes production-quality Python with tests
- AI-native development: works daily with coding agents that read the codebase, run tests and iterate on their own, while you set the scope and review the diffs. Can explain and correct what the agent produced
- Clear written communication in design docs and PR descriptions
Preferred Qualifications
- Has shipped an LLM or agent application to real users in production
- Has run coding agents with more autonomy — letting them carry whole features end to end under constraints you set, or building pipelines where agents build, test and ship with humans designing the system
- Agent memory: persisting and retrieving user or task context across sessions
- Feedback loops: turning user feedback and production traces into improved skills, tools or prompts, with human review
- Context engineering: progressive disclosure, context management for long multi-turn sessions, prompt caching
- Lightweight decision models (e.g. Jev) for routing, intent classification or caching
- Guardrails and verification for agents that act on enterprise data
- Background in predictive modeling, segmentation, or experiment design; understands correlation vs causation
- Experience with Databricks or Spark; LLM observability tools (e.g. Langfuse)
- Master’s or PhD degree
- Industry experience in MarTech and passion for building customer-centric products
#LI-TC1
How we rate this
(Senior) Machine Learning Scientist (LLM & Agents) at Appier rates 94 out of 100 for how much of the daily work is AI. That makes it Builds AI (AI Level 4 of 4). The level is about AI in the job, not seniority.
Builds AI. The job is building AI systems.
- ●●●● Builds AI80 to 100
- ●●●○ Works on AI60 to 79
- ●●○○ Uses AI40 to 59
- ●○○○ Little AI0 to 39
Levels come from how often the tools, models and workflows of the role are named in the posting itself. Read the description and count the AI tools, models and workflows it names.
Prepare for this job
Built from this posting and from similar jobs on Level.
Skills and AI tools this role asks for
Questions you could be asked
- How do you decide when an AI agent can act on its own versus asking for approval first?
- Walk me through how you've used OpenAI in your day-to-day work.
- What are the limits of Claude that you've run into, and how did you work around them?
- What's a project where you used Databricks hands-on?
- Walk me through how you've used Codex in your day-to-day work.
Adapt your resume
- List these exact terms on your resume: AI agents, OpenAI, Claude, Databricks, and Codex. An applicant tracking system matches the wording, not the idea.
- Attach one line of real, concrete experience to at least one of them. A tool named with nothing behind it rarely survives a human read.
- Lead with what you built, trained or shipped. This role is judged on the AI system itself, not the tools around it.
Want an expert to read your CV for this job?
Free. Send your CV and the role you want next. You get your review by email within 24 hours.
Get new data jobs (Builds AI ●●●●) by email
One email a week with the new data jobs (Builds AI ●●●●), each rated for how much AI is in the work. No recruiter spam, unsubscribe in one click.
Free. One email a week. Unsubscribe in one click.
Similar roles
Data roles that build AI, at other companies.
What kind of AI work fits you?
Answer 12 practical questions in about three minutes. Get a simple profile, the work it points to, and live roles to explore next.
Find my next step