Data Scientist, Analytics
At a glance
Mid-level Data Science role at Appier. Taipei.
Pay not stated Data Scientist salary
dataroles summary, based on the employer's posting.
What you'll do
- Analyze campaign, bidding and user activity data to find performance opportunities and turn them into testable ideas
- Investigate shifts and conflicting metrics until you identify the underlying cause
- Work with ML, product, creative, engineering and operations teams to move improvements into live campaigns
- Design and assess online A/B tests, set success measures and guardrails, and make launch decisions
- Build metric frameworks, datasets, pipelines and dashboards so teams can track campaign health and act on results
What you bring
- At least three years in data analytics or data science in a product- or performance-focused setting
- Strong experimental statistics, including hypothesis tests, power calculations and interpretation of noisy online tests
- Python, advanced SQL and large-scale data processing with PySpark or a comparable distributed framework
- Working ML knowledge, clear communication across technical and non-technical groups, and fluent spoken and written English
Who this fits
This role suits a curious analyst who checks results carefully, uses AI tools in daily work, and can influence teams without direct authority. You’ll take initiatives from analysis through implementation, collaborating across technical and operational teams.
From the employer
About Appier
Appier is an AI-native Agentic AI as a Service (AaaS) 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
Appier's Ad Cloud handles millions of bid requests per second, and that number keeps climbing. Behind every request is a chain of decisions: how much to bid, what to recommend, which creative to show, how fast the system responds. Each one is a lever on campaign performance, and at this scale a 1% improvement is enormous.
Your job is to find those levers in the data. You will turn campaign, auction, and user-behavior data into insights about where the next performance gain is, prove it with rigorous A/B testing, and drive the change across the organization. This is not a back-office role. You will work hand-in-hand with ML scientists, engineers, product managers, and the operations team, and you are expected to set the agenda with data and push until the improvement shows up in live campaigns.
AI now handles much of the mechanics: writing queries, drafting code, producing charts. What it does not do is decide which question is worth asking, notice that a number does not add up, or keep digging when the first explanation is too convenient. That is the job. We are looking for someone who is relentlessly curious about why, and who cannot leave an unexplained anomaly alone.
What You Will Do
- Find where performance comes from. Mine campaign, bidding, and user data to identify the highest-leverage opportunities to improve online campaign performance, and turn them into testable hypotheses.
- Get to the root cause. When performance shifts or metrics contradict each other, dig until you understand the real mechanism, not just the correlation.
- Drive improvements across functions. Partner with ML scientists on bidding and recommendation algorithms, with product and creative teams on creative serving, with engineering on system improvements, and with operations on how campaigns are run. Your insights set the direction; you follow through until the impact lands in production.
- Own experimentation. Design, run, and read a high volume of A/B tests on live campaigns. Define metrics and guardrails, diagnose noisy results, and make clear ship / no-ship calls.
- Build the foundation and make it visible. Define metric frameworks, build the datasets and pipelines behind them, and deliver dashboards that let stakeholders monitor campaign health and act on findings without waiting for a report.
- Work with AI, not just on it. Use AI assistants and agentic tools throughout your workflow so that the team's analytical throughput keeps growing.
You Will Thrive Here If
- You ask "why" one more time than is comfortable, and you verify what AI (and people) hand you.
- You influence scientists, engineers, PMs, and operations teams without direct authority, and you own an initiative from insight to shipped result.
- You treat AI tools as a daily multiplier, with the judgment to know when their output is wrong.
About you
[Minimum qualifications]
- 3+ years of hands-on experience in data analytics or data science in a product- or performance-driven environment.
- Strong statistical foundation in experimental design: hypothesis testing, power analysis, and reading noisy online experiments.
- Proficiency in Python and expert-level SQL, with hands-on experience processing large-scale data in PySpark or a comparable distributed framework.
- Working knowledge of machine learning sufficient to collaborate with ML scientists and evaluate models' real-world impact.
- Demonstrated ability to turn analysis into decisions and communicate clearly with technical and non-technical audiences.
- Fluent English, written and spoken.
[Preferred qualifications]
- Experience in programmatic / digital advertising (RTB, DSP, ad networks) and its core metrics (CTR, CVR, CPA, ROAS, win rate).
- Experience with causal inference or building experimentation infrastructure.
- Experience building dashboards (Looker, Tableau, Metabase, Grafana, Superset, or code-based libraries).
- Experience building automated or agentic analytics workflows with LLMs.
Data Science work from LinkedIn, X, Reddit and Upwork, in one list.
Each morning you get everything that fits, strongest first.
Data Science
Find me workBy the team behind dataroles.