Tho Doan

Data Scientist with 3+ years of experience developing and deploying machine learning solutions across demand forecasting, supply chain analytics, and financial services. Proven track record of building production-grade ML algorithms — from feature engineering and model development through testing, deployment, and performance monitoring — with a strong focus on real-world business impact. Experienced collaborating across Data Science, Engineering, and Product teams to deliver solutions aligned with product roadmaps.

Deep expertise in Python (scikit-learn, pandas, NumPy, scipy, XGBoost), SQL, and MLOps tooling. Solid foundations in statistical inference, probability, Bayesian reasoning, and supervised/unsupervised learning applied to forecasting, segmentation, and optimisation problems.

Data Analyst at Financial Services Company (2026-03 – Present)

• Designed and built ML-based analytical models covering agent productivity, revenue attribution, and funnel conversion across 5+ financial product lines — translating complex business requirements into end-to-end data science solutions.
• Developed statistical customer funnel analysis models applying hypothesis testing and segmentation to surface actionable commercial insights proactively.
• Delivered a C-level Power BI dashboard with automated reporting pipelines (same-day data freshness), reducing report production from 3+ days to under 2 hours.
• Defined both technical (AUC, precision/recall) and business KPI metrics to measure solution effectiveness, with ongoing post-deployment performance monitoring.

Data Business Analyst at THACO GROUP (2025-01 – 2026-02)

• Led end-to-end ML problem framing and algorithm development for cost management analytics across Finance, Accounting, Production, and Logistics.
• Built ML-based data reconciliation and anomaly detection framework across 10+ heterogeneous source systems (ERP, SAP, SCM), improving data quality SLA from ~85% to 98%+.
• Developed forecasting and cost performance models (Budget vs. Actual vs. Forecast, KPI logic, cost allocation) for C-level leadership; defined business and technical metrics to evaluate model effectiveness.
• Maintained full model development lifecycle documentation — problem definition, feature logic, decision points, result interpretation — ensuring team knowledge transfer.
• Ran A/B-style testing of analytical models with business units; evaluated performance during rollout and iterated based on metric feedback.

Data Analyst at THACO GROUP (2025-03 – 2025-12)

Project Resource Management Analytics

• Applied statistical analysis and ML segmentation to resource allocation data across 5+ concurrent IT projects — detecting utilisation bottlenecks and capacity risks proactively.
• Delivered resource performance dashboards and analytical reports for Project Managers; presented recommendations to senior stakeholders in non-technical language.

Data Analyst at FPT Software (2023-12 – 2024-12)

• Translated complex business requirements from 5+ enterprise clients into analytical data models and ML feature logic, reducing downstream report errors.
• Contributed to presales data science scoping for 4+ proposals — defining algorithmic problem statements, ML methodologies, and expected business impact metrics.
• Applied pattern detection and statistical analysis to identify reporting anomalies; presented findings to non-technical client stakeholders with clear analytical narratives.

Data Analyst — Manufacturing & Supply Chain at Wrapex Ltd (Prowrap) (2022-01 – 2023-12)

• Applied statistical pattern detection and segmentation to large operational datasets, identifying 15%+ sourcing inefficiencies that shaped financial and procurement strategy.
• Built inventory and supplier cost analytics models supporting cost optimisation decisions for senior management across manufacturing and supply chain operations.

Data Operations Analyst at Title Insurance Company (2024-08 – 2025-02)

• Developed SQL-based anomaly detection scripts across 3+ regulated financial systems, achieving 99%+ processing accuracy and reducing manual QA effort by 60%.
• Applied statistical validation methods to high-volume property transaction data, establishing data quality benchmarks for downstream ML models and reporting pipelines.

MSc in Data Science – University of the West of England (2021-09 – 2022-09)

BEng in Electronics Engineering – University of the West of England (2019-09 – 2021-09)