Job Description
Location: Bali in office
About Us
TyrAds is a leading tech-driven loyalty and rewards platform that partners with businesses to create meaningful and rewarding experiences for their customers. With millions of users across our platforms, we specialize in innovative AdTech solutions powered by big data, machine learning, and deep learning technologies.
Our systems process over 10,000 events per second, delivering real-time insights at scale. We build seamless user experiences across web, Android, and iOS, while leveraging ML/AI to power personalization and advertising effectiveness.
We are a team of 90+ employees worldwide and growing, guided by values of transparency, ownership, learning from mistakes, and respect for diverse cultures. Our product development follows Agile methodologies with weekly releases and collaboration tools such as GitHub, Jira, and Slack.
We're hiring a Data Scientist who can ship real-time ML systems and also evaluate and improve them after launch. You'll work on scoring models, recommendations, fraud/risk signals, and the data pipelines behind them.
Job Requirements
Must Have
- Production ML
- Deployed real-time ML models behind an API
- Debugged latency issues, optimized inference
- Familiar with monitoring/observability for ML systems
- Technical
- 3+ years in DS/ML
- Strong Python (pandas, numpy, scikit-learn)
- Strong SQL (CTEs, window functions, large tables)
- Classification models (XGBoost/LightGBM) and evaluation (AUC, precision/recall)
- Analytics
- A/B testing experience (design, analysis, avoiding bias)
- Can communicate insights clearly and drive decisions
- Engineering
- Git + Docker basics
- Clean, maintainable code
Nice to Have
- Fraud detection
- Databricks / Spark
- Feature stores, MLflow, or similar
Job Responsibilities
What You'll Do
- Analytics & Iteration
- Investigate funnel and performance issues, find patterns, propose fixes
- Design and analyze experiments (A/B tests, holdouts)
- Build post-launch evaluation
- Turn findings into clear recommendations for product/engineering stakeholders
- Real-time ML in Production
- Build and deploy real-time scoring/recommendation models (not just notebooks)
- Work with latency constraints (optimize p95/p99, payload size, feature fetch time)
- Implement robust serving
- Ensure training/serving consistency
- Monitoring & Model Health
- Set up monitoring for live models: data drift, prediction drift, performance decay
- Define guardrails and alerts
- Recommend improvements (features, thresholds, retraining cadence)
