Kiến trúc hạ tầng dữ liệu quy mô lớn, xây dựng Feature Store và nền tảng AutoEDA phục vụ hệ sinh thái Zalo.
Yêu cầu
5+ năm kinh nghiệm thực chiến với hệ sinh thái Big Data (Spark, Kafka, ClickHouse...)
Kinh nghiệm với Open table formats (Iceberg, Delta, Hudi) và tối ưu lưu trữ, tính toán
Thành thạo Scala hoặc Python xử lý dữ liệu hiệu năng cao, SQL
Kinh nghiệm pipeline streaming/batch, CDC Debezium, Medallion Lakehouse, dbt
Kinh nghiệm xây dựng API gRPC/REST trên K8s, Docker, Airflow, CI/CD, Observability
Trách nhiệm
Xây dựng và tối ưu pipeline dữ liệu cho khai thác dữ liệu, nhắm mục tiêu và phân tích, tuân thủ quy định bảo vệ dữ liệu
Thiết kế pipeline đồng bộ dữ liệu giữa các cơ sở dữ liệu, hỗ trợ xử lý batch + stream và phục vụ online độ trễ thấp
Xây dựng nền tảng analytics self-serve (AutoEDA) để các bên liên quan khai thác dữ liệu
Thiết lập best practices kỹ thuật; cố vấn, code review và xây dựng văn hóa kỹ thuật cho đội ngũ
Có thì tốt
Kinh nghiệm quản trị HDFS, Object Storage: backup, cleanup, capacity planning
Kinh nghiệm xây dựng Feature Store tập trung quy mô lớn
Hiểu biết về AutoML, MLOps, quy trình Data Science, DMP/CDP, Vector DB
Tóm tắt bằng AI từ JD gốc, đọc JD gốc trước khi ứng tuyển.
Build and optimize data pipelines for diverse use cases (data mining, audience targeting, analytics) with strict compliance to data privacy regulations;
Engineer reliable pipelines to synchronize data across databases, powering batch + stream processing, large-scale analytics, and low-latency online serving;
Collaborate with Data Science, Policy, System Operation, and Research teams to understand their data needs and deliver solutions;
Establish solid design and engineering best practices for both technical and non-technical partners.
Technical skills:
5+ years hands-on in the Big Data ecosystem (Spark, Kafka, ClickHouse, Cassandra or similar);
Open table formats (Apache Iceberg / Delta / Hudi) with an open catalog (Unity / Polaris / Glue); storage & compute optimization (compression, partitioning, bucketing, ORC/Parquet, compaction, small-file handling);
Reliable pipelines syncing data across databases; CDC (Debezium), Kafka; unified batch + stream processing with exactly-once/idempotency, schema evolution, backfill;
Data contracts, testing, SLA/SLO, data observability & lineage;
Build the self-serve, AI-backed AutoEDA / analytics platform enabling stakeholders to mine data without engineering overhead (serving DS/ML — model training not required);
API & Platform: High-throughput APIs (gRPC, REST) on Kubernetes; latency & scalability optimization;
AutoML & MLOps: auto-training, evaluation, prediction at scale; model registry & serving; end-to-end ML pipeline understanding;
Fundamental knowledge of data science workflows and ML integration;
Experience with DMP/CDP and vector databases for AI/RAG use cases.
Join the Audience Platform team at Zalo Group and architect the data backbone for one of Vietnam's largest digital ecosystems. With nearly 80 million Monthly Active Users and TBs of data flowing through our systems daily, we're looking for a Data Engineer to build robust pipelines that power large-scale data mining and give decision-makers seamless access to actionable insights.
Scope of work: You won't just move data - you'll build the platforms that make data mining effortless. You will develop a centralized Feature Store and self-served, AI-backed automated analytics (AutoEDA) serving the Ads, Fintech, and VAS business lines. Your mission is to bridge massive datasets and business impact through thoroughly architected, highly optimized, rock-solid data infrastructure.
Join the Audience Platform team at Zalo Group and architect the data backbone for one of Vietnam's largest digital ecosystems. With nearly 80 million Monthly Active Users and TBs of data flowing through our systems da...
