📍 Work Location: 98 Hoang Quoc Viet, Cau Giay, Hanoi
🕒 Working Hours: 8:30 AM - 5:30 PM (Monday to Friday)
💰 Salary Range: VND 30-70 million
What You'll Do:
- Design and operate scalable, secure ML platform infrastructure supporting offline training, batch scoring, and real-time inference.
- Build and maintain core ML platform services such as model registry, versioning, champion/challenger workflows, feature ingestion and serving, and deployment orchestration.
- Standardize ML workflows from experiment → approval → production.
- Build and maintain CI/CD pipelines for ML and platform components, including automated deployment, environment promotion (UAT → Production), rollback, and recovery.
- Deploy and operate model serving systems via REST / gRPC APIs and batch scoring pipelines.
- Support advanced deployment strategies such as shadow, canary, and A/B testing.
- Ensure low latency, high availability, and strong observability for scoring services.
- Collaborate with Data Engineering teams to build and operate offline and online Feature Stores, ensuring feature consistency between training and serving.
- Design and implement monitoring for model performance, data quality, feature stability, and system health.
- Build early-warning mechanisms for model decay, data drift, and production anomalies.
- Support model governance, auditability, and compliance in credit-risk contexts.
- Work closely with Data Scientists, Risk, Fraud, Backend, and Infrastructure teams.
- Enable Data Science teams to move from notebook to production safely.
What We're Looking For:
- Proven experience as an ML Platform Engineer, MLOps Engineer, or Platform / Software Engineer working with ML systems.
- Strong Python skills for platform services, automation, and ML tooling.
- Solid understanding of the ML lifecycle in production environments.
- Experience with containerization and orchestration (Docker, Kubernetes).
- Experience building and operating production APIs (FastAPI, gRPC).
- Strong understanding of batch and real-time scoring systems.
- Familiarity with CI/CD pipelines and infrastructure-as-code mindset.
- Strong system-thinking, with ability to design scalable, observable, and governed platforms.
- Excellent communication skills and strong ownership mindset.
Tech Stack (at a glance)
- Data & Storage: MinIO, Iceberg, ClickHouse
- Processing: Spark, Polars, Airflow, Pathway
- ML & MLOps: [protected info], MLflow
- Serving: FastAPI / gRPC
- Platform: Docker, Kubernetes, CI/CD (Git-based)
- Monitoring & Governance: Prometheus, Grafana, model & data observability
Nice to Have (Bonus)
- Experience in credit scoring, risk, fraud, or financial services.
- Experience with Feature Stores and Lakehouse-style data platforms.
- Experience with streaming or real-time data systems.
- Knowledge of model governance, approval workflows, and A/B testing for
- decision systems.
Why Join VC Scoring Team
- Build the core ML platform powering real financial decisions.
- Work on production-grade ML systems with high impact and responsibility.
- Own architectural decisions and influence long-term platform strategy.
- Collaborate with strong Data, Risk, and Engineering teams.