About You
- You're a go-getter with mad juggling skills (or multiple hats) who can thrive in a fast-paced, agile environment
- You enjoy doing purpose-led and meaningful work
- You have a strong thirst for knowledge and are driven to find solutions that don't exist yet
- You are comfortable with ambiguity and extremely resourceful (in your past life, you could've been a detective)
- You always find a way to get things done without sacrificing the quality of your work, integrity, and values
- No task is off limits for you
- You are humble and prioritize the success of the team over your own with an eagerness to help those around you
- You don't shy away from challenges and can bounce back from setbacks
- We strongly encourage individuals with disabilities to apply. We believe in equal opportunity and strive to create an inclusive workplace where everyone can thrive.
What you'll do and what success looks like in this role:
Analytics, Experimentation & Statistics
- Use advanced SQL / BigQuery for data exploration, feature engineering, and large-scale analysis.
- Build and maintain BI dashboards/reporting (Looker / Looker Studio, Tableau, or Power BI).
- Design and analyze end-to-end A/B tests: sample size determination, experiment design, pre/post analysis, and statistically valid data-driven recommendations.
- Apply strong statistical reasoning to validate findings and measure uncertainty.
- Manage mobile-app event tracking pipelines end-to-end — instrumenting user activity within the app (Customer Experience) via Firebase Analytics, streaming events through Google Analytics (GA4), and storing them in BigQuery via GA4 export — ensuring all user activities are reliably captured, modeled, and available for analytics and downstream ML features.
Machine Learning & Modeling
- Design, build, validate, deploy, and maintain production ML models across various problem types: propensity/credit-risk scoring, churn/retention prediction, behavioral segmentation, and product recommendation systems, applying MLOps principles (MLflow / Vertex AI Experiments).
- Develop gradient boosting models (LightGBM / XGBoost) and regression models for binary classification and scoring; conduct rigorous evaluation using AUC, IV, GINI, KS, precision/recall, and out-of-time validation; monitor drift (PSI) and perform recalibration or scheduled retraining.
- Build and fine-tune transformer / NLP models for text classification, including short-text and multilingual (Indonesian) use cases.
- Develop recommendation and personalization systems (collaborative filtering, ensemble k-NN, ranking/multi-output models) serving large-scale user bases.
MLOps, Pipeline & Deployment
- Orchestrate end-to-end ML pipelines in Vertex AI / Kubeflow Pipelines (KFP) — training, batch inference, APIs, scheduling, and parameterization.
- Containerize and deploy services (Docker, Cloud Run) and manage CI/CD (GitLab CI or equivalent).
- Manage GPU training jobs, quotas, reproducibility, and model/artifact versioning.
- Write clear runbooks and documentation to ensure models and pipelines can be operated and maintained by broader teams.
GenAI & Intelligent Automation
- Build LLM-based workflows (e.g., Google ADK / LangChain) for automated insight generation, summarization, and conversational analytics (RAG-style chatbots on internal data), including maintaining internal websites (TypeScript) and their backend (FastAPI).
- Automate periodic analytics reporting (daily/weekly/monthly) and stakeholder notifications via channels such as Microsoft Teams and email, including auto-generated documents and slide decks.
Stakeholder Engagement & Communication
- Translate complex data findings into clear, actionable recommendations for non-technical stakeholders.
- Partner with product, marketing, business, and engineering teams; mentor junior analysts and interns as needed.
What Is Required and What We're Looking For
- Bachelor's or Master's degree in Computer Science, Statistics, Mathematics, Quantitative Science, Business Analytics, or a related field (Master's preferred).
- Proven experience of at least 5–7 years as a Senior Machine Learning Engineer / Data Scientist in a product/production environment — with models that have actually been deployed to production, not just prototypes or notebooks.
- Strong proficiency in Python (pandas, NumPy, scikit-learn, LightGBM/XGBoost, PyTorch, TensorFlow/Keras), SQL/BigQuery for data exploration/analysis, and NoSQL such as MongoDB/Firestore.
- Hands-on experience with cloud-based ML platforms — preferably Google Cloud (Vertex AI, BigQuery, Cloud Run, Storage, Compute Engine) — as well as pipeline orchestration (Kubeflow Pipelines or similar).
- Demonstrated experience in building behavioral segmentation (RFM models, K-Means, and other clustering techniques), predictive analytics/scoring, recommendation engines, and A/B testing mechanisms.
- Proven experience managing end-to-end mobile app event tracking — Firebase Analytics → Google Analytics (GA4) → BigQuery export — to capture, structure, and model in-app user activity.
- Skilled in designing and running A/B experiments, including sample design, pre/post analysis, and developing monitoring dashboards.
- Strong foundation in statistics and the ability to apply it in analysis.
- Excellent written and verbal communication skills, with the ability to convey complex insights to diverse, non-technical audiences.
- Ability to work effectively in collaborative, cross-functional teams; flexible and adaptable to evolving project needs; highly organized and detail-oriented.
Skills Mandatory
- Proficient in Python and core ML/data stack (pandas, scikit-learn, LightGBM / XGBoost, PyTorch, TensorFlow/Keras).
- Advanced SQL / BigQuery skills.
- Experience with cloud-based ML platforms, preferably GCP / Vertex AI.
- Version control using Git, including built-in CI/CD workflows.
- Familiarity with microservices architecture and containerization (Docker / Podman).
- Strong understanding of clustering, classification, and scoring techniques, as well as A/B testing.
- Ability to design, develop, and maintain full-stack web applications, including frontend, backend, and supporting utilities.
- Experience integrating systems with mobile applications and third-party vendor services.
Skills Preferred / Nice to Have
- MLOps tools: Kubeflow Pipelines / Vertex AI Pipelines, Docker, Cloud Run, GitLab CI/CD.
- Deep learning / NLP: Hugging Face Transformers, PyTorch; experience with encoder model fine-tuning.
- GenAI / LLM: Google Gemini (or OpenAI / Anthropic), retrieval-augmented generation (RAG), prompt engineering, and LLM application development.
- Experience with recommendation systems, Customer Lifetime Value (CLV) models, and ranking models.
- BI tools: Looker / Looker Studio, Tableau, Power BI.
- Product analytics/event instrumentation: Firebase Analytics, Google Analytics (GA4), and GA4-to-BigQuery exports.
- Experiment tracking and model registry tools (e.g., MLflow, Vertex Model Registry).
- Experience with distributed or large-scale data processing (PySpark or BigQuery ML).
- Background in financial services, banking, or fintech industries.
- Familiarity with credit risk or fraud modeling, including model governance and monitoring in regulated environments.
- Experience handling PII/sensitive data with appropriate encryption and compliance controls.
- Experience mentoring junior team members or leading small-scale technical initiatives.
