About the job
About Tencent
Tencent is an Internet-based platform company founded in Shenzhen, China, in 1998. We use technology to enrich the lives of Internet users and assist the digital upgrade of enterprises. Our mission is Value for Users, Tech for Good. We embrace a culture of teamwork & creativity and are driven by our values - Integrity, Proactivity, Collaboration and Creativity.
We are rapidly expanding our international operations and are looking for top talent to propel us forward. Combining the results-oriented nature of a start-up with the resources of a profitable and leading Internet company, Tencent offers a unique opportunity for aspiring individuals to thrive.
About the Business
Interactive Entertainment Group Global (IEGG) is responsible for the Global R&D, operation, and development of the company's interactive entertainment business including games and eSports. Through online gaming, live broadcasts and offline eSports, IEG assists the company in leading the global interactive entertainment market to create better interactive entertainment content experiences for users
Job Description
We are seeking a core partner to co-own our next-generation Anomaly Detection & Attribution (ADA) platform. This platform powers real-time monitoring, automated diagnostics, and decision support across multiple games and regions. You will take ownership of the full lifecyclefrom architecture and development to maintenance and analytical impactwhile advancing LLM-Agent capabilities that convert noisy signals into clear, actionable narratives for business teams.
Key Responsibilities
- Platform Ownership: Design, implement, and maintain robust ADA pipelines for T+1 or near-real-time anomaly detection, multi-baseline benchmarking (global, regional, and country-level), and multi-source attribution (holidays, version changes, events, migrations, user behavior).
- Detection & Inference: Productionize methods for change-point detection, outlier analysis, time-series feature engineering, causal/ablation checks, and automated storyline generation explaining what happened, why, and recommended actions.
- AI Agent Development: Build AI-driven agents using tool-use and reasoning flows (ReAct, LangGraph, or similar) to enable conversational drill-downs for operations and product teams.
- Insights Productization: Deliver dashboards, alerts (Email/Chat/WeCom), and concise decision memos. Collaborate with stakeholders in publishing, operations, marketing, and analytics to iterate and refine outputs.
- Collaboration & Documentation: Work closely with Data Science, Engineering, and Product teams to scope, roadmap, and ship features. Maintain architecture diagrams, APIs, and operational runbooks.
Required Qualifications
- Experience: Minimum 2 years as a data analyst or in a similar data-focused role.
Technical Skills:
- Strong Python engineering skills (clean code, testing, packaging).
- Solid SQL proficiency for large-scale analytical workloads.
- Hands-on experience with time-series anomaly detection (change-point, robust statistics, seasonality/holiday adjustment, multivariate signals) and attribution logic.
- Practical exposure to LLM applications, including tool/function calling, retrieval/RAG, and agent planning, with at least one framework/API (e.g., OpenAI, Claude, DeepSeek, LangChain, LangGraph).
Mindset & Ownership:
- Systems and product thinking: ability to translate business challenges into measurable detection/attribution logic.
- Strong ownership and reliability: implement guardrails, monitors, and documentation; debug in production and prevent regressions.
Preferred / Nice-to-Have
- Model evaluation & prompt engineering: rubric design, offline evaluation sets, golden tasks, prompt/test versioning, data flywheels.
- Causal inference & experimentation: diff-in-diff, CUPED, synthetic controls, online A/B testing at scale.
- Dashboarding & alerting experience: Superset, Tableau, Looker, Metabase; alerting via Slack/WeCom/Email with noise-reduction heuristics.
- Domain expertise in gaming analytics: retention funnels, reactivation, event/version rollout attribution, fraud/smurf detection.
- Infrastructure & operations: Docker, Kubernetes, CI/CD, IaC; cloud platforms (GCP, AWS, Azure); cost and performance tuning.
- Additional development experience is a bonus.
