I am a data scientist with 10 years in financial services and consulting, working across risk, fraud and trust. Recent work: cut multi-document review effort with a RAG framework. Also lifted detection precision via classical ML techniques for emerging risk themes by double digits, shifting monitoring from quarterly to daily.

I specialise in translating ambiguous, open-ended business problems into production-grade data science solutions, owning the lifecycle from exploratory analysis to experimental design (Hypothesis Framing, AB Testing, Shadow Testing). This spans imbalanced datasets, behavioural networks, and propensity models, with decision engines built in Python and SQL using Scikit-Learn and tree-based ensembles (Random Forest, XGBoost), plus graph analytics for network behaviour. I extend this with GenAI (RAG, agentic workflows), causal inference, and NLP/LLMs to make sense of behavioural and unstructured risk signals.

For the last 2+ years, I’ve been a hands-on technical lead working as a “player-coach” by architecting solutions, writing production code, and driving cross-portfolio (xPA) data science initiatives with a pod of 5+ (data scientists, interns, and apprentices), partnering with business and product stakeholders and translating technical work into business impact. This includes shaping and setting department-wide data science standards.

• 🌍 I'm based in Bangalore, India
• ✉️ You can contact me at monu.singh9203@gmail.com
• 🤝 I'm open to collaborating on Data Science / ML Projects, Kaggle Competitions, Coaching & Training, Consultancy