[Intern] Big Data Development Engineer Intern
Key Responsibilities:
Assess the Status Quo — Develop an end-to-end understanding of the campaign lifecycle and available data assets; identify all decision points currently reliant on manual judgment (opportunity identification, audience selection, budget estimation, campaign mechanics design, post-campaign review) and determine where intelligent automation yields the highest return.
What Campaign to Run — Mine campaign opportunities from user behavior, product and market signals, and historical campaign performance; build an AI Agent that outputs candidate campaign themes with clear supporting evidence, not just a ranked list.
Who to Target — Build or reuse user segmentation and response/propensity models; apply uplift modeling to direct budget toward truly incremental audiences who would be influenced, rather than users who would convert anyway.
How Much Budget — Build a cost-benefit model: projected participation, incremental trading volume / revenue, ROI simulation, budget allocation recommendations, and sensitivity and risk analysis (including abuse and reward-farming risks under incentive designs).
Feasibility Analysis & Plan Design — Translate model outputs into a reviewable proposal: objectives, mechanics, target audience, budget, expected KPIs, and risk control boundaries; align with PM, Growth, and business stakeholders.
MVP Delivery — Implement the end-to-end pipeline (data → model → recommendation → campaign configuration → A/B or holdout evaluation); run it on at least one real campaign or rigorous backtest; quantify improvement relative to the current manual decision-making baseline.
Requirements:
Undergraduate or graduate student in Computer Science, Data Science, Statistics, Economics, or a related field.
Solid proficiency in SQL and Python. Experience with real-time / streaming tech stacks (Flink, Kafka) is a plus.
Data science foundations relevant to this project: A/B experiment design, causal inference and uplift modeling, propensity/response models, and basic predictive modeling. Must be able to distinguish correlation from incremental effect.
Hands-on experience with LLM application development: prompt engineering, RAG, Agent / tool-calling frameworks; able to use LLMs for reasoning and orchestration on top of quantitative models.
Business acumen: genuine interest in fintech / crypto and growth marketing; able to translate a business question ("Should we run a trading competition next month?") into a data problem, and translate the conclusion back into an actionable decision for business stakeholders.
Strong communication skills — this role sits at the intersection of data, PM, and growth/marketing.
Bonus: experience in growth / CRM / campaign analytics, recommendation systems, or marketing mix modeling background.
Minimum 3-month internship commitment, 5 days per week on-site.
Fluent in Mandarin is required.