Data & AI Engineering Intern

【 About the role 】
Our centralized Data Team is building an Agentic Data Ecosystem: a unified semantic layer, a User 360 foundation, and AI agents that help teams across the company get answers from data without waiting on an analyst.
As a Data & AI Engineering Intern, you'll work directly with our data engineers, analytics engineers, and data scientists on three hands-on areas. You'll use AI agents to automate data collection, keep our ingestion pipelines healthy, and help make our internal Analytics Agents more accurate. This isn't a "make coffee and watch" internship. What you build will be used by marketing, growth, and product teams across our global offices.
【 What You'll Do 】

• AI-powered data collection: Use AI tools and agents (e.g. Claude, browser automation, scheduled workflows) to set up recurring collection of marketing data, including from advertising and e-commerce platforms that don't offer a public API.
• Data ingestion & cleaning pipelines: Maintain and monitor ingestion and cleaning pipelines in Databricks. Investigate failed jobs, fix schema changes caused by upstream platform updates, and help catch data discrepancies before stakeholders do.
• Agent answer quality: Review and tag question–answer pairs from our internal Analytics Agents, build evaluation sets, and help fine-tune prompts, semantic definitions, and knowledge sources to improve answer accuracy.
• Documentation: Write clear runbooks and notes for the pipelines and agent workflows you work on, so the team can build on them after your internship.

【 Our Stack 】

• Databricks on AWS (lakehouse, pipelines, Genie / Agent Bricks)
• Python and SQL
• Amplitude (product and marketing analytics)
• Tableau (stakeholder reporting)
• Claude and other AI agents / MCP-based tools
• Git-based version control

【 Required Qualifications 】

• Currently pursuing (or recently completed) a degree in Computer Science, Computer Engineering, Information Management, or a related field.
• Hands-on experience using AI tools (e.g. Claude, ChatGPT, Cursor, GitHub Copilot) to build or automate something real, not just for chat. Please share an example in your application.
• Working knowledge of Python, and basic SQL.
• Strong learning ability: comfortable picking up new tools and platforms quickly, and figuring things out from documentation.
• Attention to detail and a habit of checking whether data "looks right."
• Able to commit 2–3 days per week for at least 3 months.
• Good communication in Mandarin and English (reading and writing technical English is essential).

【 Preferred Qualifications 】

• Experience with web scraping or browser automation (e.g. Playwright, Selenium, Puppeteer).
• Exposure to ETL/ELT concepts, data pipelines, or cloud data platforms (Databricks, Spark, BigQuery, Snowflake).
• Experience building or evaluating LLM applications: prompt engineering, RAG, agent frameworks, or LLM evaluation.
• Familiarity with digital advertising platforms (Meta, Google, TikTok, Amazon Ads).
• Personal projects or GitHub repos that show how you build and learn.

【 What You'll Gain 】

• Real ownership of production data work, with mentorship from an experienced data team.
• Hands-on experience applying AI agents to data engineering and analytics, a skill set most companies are only starting to build.
• Exposure to a global consumer IoT + subscription business across marketing, growth, and product.
• An opportunity to extend up to 12 months based on performance and team needs.

【 Perks 】

• Work with an international team at our Taiwan HQ on problems with global impact.
• An open, transparent working environment where your ideas are heard.
• Pet-friendly office: bring your fur babies to work!
• A LOT of free snacks and drinks for humans and furkids.
• Afternoon tea twice a month and monthly birthday parties.

【 Why Join Us? 】

• See how a modern data team uses AI agents in production, not just in demos.
• Build things that marketing and product teams across the world will use.
• Join a collaborative, high-trust team that cares about quality and learning.

If you love building with AI, learn fast, and want to turn messy real-world data into something useful, we'd love to hear from you!