HUNG LE NHAT / LEAD DATA SCIENTIST / UK

Intelligence
for financial
decisions.

I work at the intersection of artificial intelligence, financial crime and quantitative risk — turning complex data into robust, explainable models and better decisions.

CURRENTLY / LEAD DATA SCIENTIST · FINANCIAL CRIME · SANTANDER UK
⌘ FINANCIAL INTELLIGENCE ENGINE
RESEARCH VIEW / V.01
BEHAVIOURAL NETWORK / SYNTHETIC● PATTERN DETECTED
GRAPH_SCAN COMPLETE
PATTERN_TYPE CONNECTED CLUSTER
MODEL_INPUT ILLUSTRATIVE ONLY
AI/ML
AML
QUANT
● DEMONSTRATION ENVIRONMENTNO CUSTOMER OR PRODUCTION DATA
600MApprox. retail transactions / month¹
400K+Applications & customers / day²
50→78%Fraud detection accuracy · internship³
ML4HResearch publication · 2023
01 / Areas of focus

Applied intelligence.
Measurable decisions.

Combining statistical rigour with production-minded AI engineering — across finance, regulated modelling and real-world systems.

01
⌁

Financial Crime AI

Transaction monitoring, fraud detection, anomaly detection, behavioural profiling and explainable model design.

02
◈

Credit Risk & Governance

PD / LGD / EAD, IFRS 9, Basel frameworks, challenger models, validation and stability monitoring.

03
∿

Quantitative Finance

Risk factor analysis, Monte Carlo simulation, VaR, scenario design and interest-rate / option modelling.

04
⌘

ML Engineering

Scalable data pipelines, experiment tracking, model delivery, computer vision and natural language processing.

02 / Selected systems

From models
to impact.

A selection of professional domains and published research. Employer work is presented at a high level, without confidential implementations or data.

CASE / 01FINANCIAL CRIME
PROFESSIONAL / SANTANDER

Financial Crime Intelligence

Anomaly detection, customer-behaviour modelling and transaction-monitoring analytics across high-volume retail financial data.

Python · AWS · ThetaRayInternal / not public
CASE / 02RISK MODELLING
PROFESSIONAL / SANTANDER

Credit Risk & Challenger Models

Risk decisioning, scorecard evaluation, gradient-boosting challengers, portfolio drift and regulatory model governance.

PD · LGD · EAD · IFRS 9Internal / not public
RESEARCH / 03PUBLICATION
RESEARCH / ML4H 2023

SurvTimeSurvival

Transformer-based survival analysis for patients with multiple visits, combining time-varying covariates and synthetic data.

Transformer · SurvivalGANRead paper ↗
RESEARCH / 04DEEP LEARNING
RESEARCH / SURREY

Image Retrieval by Distillation

Similarity learning through knowledge distillation and quadruplet-loss techniques, with retrieval performance evaluation.

PyTorch · mAP · RetrievalView demo ↗
CASE / 05FORECASTING
PROFESSIONAL / VIETNAM ELECTRICITY

Energy Demand Forecasting

Time-series models using LSTM, Transformers and ensemble methods to predict short- and medium-term city demand.

LSTM · Transformer · ForecastingInternal / not public
RESEARCH / 06COMPUTER VISION
RESEARCH / SURREY

Explainable Medical Vision

Diabetic-retinopathy grading using ensembles, knowledge distillation and Grad-CAM for model interpretability.

CV · Grad-CAM · EnsemblesView demo ↗

Professional case studies describe areas of work only; no confidential data, models, code or employer systems are included.

03 / Career trajectory

Built across
industries.

From production AI and backend engineering to credit risk, financial crime and advanced modelling in banking. Detailed responsibilities reflect my CV.

PROFESSIONAL EXPERIENCE
2019 — PRESENT
Authentic company branding where available; click a logo to visit the employer’s site.

Santander UK

Lead Data Scientist — Financial Crime

Leading applied AI and statistical detection work in a high-volume financial-crime environment.

  • Lead development of ML, anomaly-detection and rule-based surveillance approaches for fraud, anti-money-laundering (AML) risk and suspicious transaction monitoring, across approximately 600 million retail transactions per month.
  • Build behavioural profiles, engineer detection features and investigate emerging transaction patterns, balancing false positives, detection coverage and interpretability.
  • Design and maintain scalable AWS data/ML pipelines with HUE, ThetaRay, Airflow, MLflow and MinIO for orchestration, storage, experiment tracking and model lifecycle management.
  • Collaborate with Financial Crime investigators, risk specialists, analysts, technology and business teams to translate investigation needs into production-ready analytical solutions.
Financial CrimeAML / FraudAnomaly DetectionAWSMLflowAirflowThetaRay
Senior Credit Risk Analyst — System Engineer

Retail risk decisioning and model governance across Santander credit products.

  • Analysed credit-risk drivers and developed/optimised decision policies across personal loans, credit cards and core banking, supporting 400,000+ applications and customers daily using PCSM, SAS, Python, SQL and XML.
  • Created Power BI performance dashboards; supported AWS data processing and migrations to improve reliability, scalability and operational resilience.
  • Worked with credit models and risk measures including PD, LGD, EAD and ECL; applied stress testing, statistical hypothesis testing, Monte Carlo simulation and champion–challenger gradient-boosting approaches.
  • Monitored Gini, KS, PSI and model drift; aligned validation and governance work with IFRS 9, Basel IRB/IRRBB, PRA/FCA requirements and SS1/23.
  • Conducted market and interest-rate risk assessments using parametric, historical and Monte Carlo VaR, scenario analysis, Black–Scholes and Hull–White modelling.
PCSMSASPythonSQLAWSIFRS 9BaselPD / LGD / EAD

Barclays UK

Artificial Intelligence Engineer — Summer Internship

Payment-fraud analytics and applied ML for risk controls.

  • Analysed transactional records across payment types to identify fraud signatures and high-impact behavioural risk indicators.
  • Engineered features, optimised existing classification models and communicated findings to product, risk and engineering teams.
  • According to my CV, the work improved fraud-detection accuracy from 50% to an average of 78%.
Fraud DetectionFeature EngineeringML Optimisation

Kyanon Digital

Artificial Intelligence Engineer

Delivered client-facing AI products and production data services.

  • Designed multi-source data acquisition across APIs and dynamic web pages, using Redis to improve extraction and downstream analysis throughput.
  • Built image-data ingestion, refinement and analysis pipelines combining computer vision and NLP for multimodal question-answering and insights.
  • Developed a document and bill recognition API with OCR / deep neural networks to turn physical bills into structured e-bills using Django REST.
  • Maintained CI/CD and Airflow orchestration; partnered with clients and engineers to translate business requirements into scalable deliverables.
Django RESTComputer VisionNLPRedisAirflowOCR

Vietnam Electricity (EVN)

Artificial Intelligence Engineer

Designed ML-assisted analytics and automation for the electricity sector; recognised as Employee of the Year 2022.

  • Developed demand forecasting and electricity-meter interpretation systems using LSTM, Transformers and time-series datasets, including short- and medium-term consumption forecasts.
  • Built Tableau dashboards with Airflow-based ETL and automated trend forecasting to support data-driven operational decisions.
  • Designed internal chatbot / question-answering systems using Rasa, pretrained NLP models and fine-tuned LLMs.
  • Worked on Hadoop / ClickHouse data architectures, API optimisation and AI/RPA-assisted template completion from reports and internal discussions.
TransformersLSTMTableauRasaHadoopClickHouseAirflow

FTECH Co., Ltd.

Artificial Intelligence Engineer

Research and applied AI engineering across payment risk, chatbots and enterprise analytics.

  • Developed data labelling and human-feedback classification workflows for internal NLP/chatbot research, with automated training and testing using Kubeflow.
  • Worked on fraud-detection modelling for online payments using transaction time-series and LSTM architectures.
  • Built an internal performance-monitoring chatbot to calculate and surface employee operational metrics.
KubeflowLSTMPayment FraudNLPChatbots

The HEINEKEN Company

Backend Engineer / Artificial Intelligence Engineer

Built industrial AI and backend systems for production and internal logistics.

  • Automated identification and tracking of production-system errors through computer vision, helping quality monitoring and operational risk reduction.
  • Optimised real-time tracking for internal transport systems, using knowledge-graph techniques to improve visibility, performance and system latency.
  • Worked across the enterprise data lifecycle — collection, processing, transformation, secure storage and API/backend engineering.
Computer VisionKnowledge GraphsBackend APIsData Engineering
04 / Research & education

Research-minded.
Production-focused.

My academic foundation spans mathematics, statistics, computer science and advanced machine learning. I translate these methods into interpretable, production-oriented systems across financial services and beyond.

CONFERENCE RESEARCH / ML4H 2023

SurvTimeSurvival:
Learning across patient trajectories.

My MSc dissertation developed a deep survival-analysis approach for patients with multiple hospital visits and longitudinal records. The work focuses on modelling time-varying covariates to improve survival-time prediction in the presence of sparse or incomplete patient histories.

The research combines a Transformer-based architecture with SurvTrace and uses SurvivalGAN for synthetic data generation. It was accepted at the Machine Learning for Healthcare (ML4H) 2023 conference, as recorded in my CV.

  • Research question: How can deep learning represent repeat visits and changing patient measurements over time?
  • Technical methods: Temporal attention, survival modelling, synthetic-data generation and evaluation of predictive outcomes.
  • Potential applications: Patient survival prediction, clinical trial planning and treatment-related research.
Longitudinal MLTransformerSurvTraceSurvivalGAN
Read the publication ↗
MULTI-VISIT TEMPORAL REPRESENTATION / ILLUSTRATIVE
PUBLIC RESEARCH & PROTOTYPES

Beyond the dissertation.

Applied research exploring representation learning, explainable computer vision and forecasting. The projects below are described in my CV; linked demos are public research artefacts.

R / 01 · 2023REPRESENTATION LEARNING

Image retrieval via knowledge distillation

Developed an image-similarity retrieval approach with supervised and unsupervised learning, transfer learning and metric-learning losses, including the proposed quadruplet-loss method. Evaluated retrieval using mean average precision and precision–recall curves.

Metric learningQuadruplet lossmAP / PR
Explore interactive demo ↗
R / 02 · 2023EXPLAINABLE MEDICAL AI

Diabetic retinopathy grading

Investigated ensemble learning, knowledge distillation and pre-trained vision models for retinal-image classification. Applied Grad-CAM to visualise areas contributing to model predictions, alongside data-quality analysis before training.

Computer visionGrad-CAMEnsembles
Explore interactive demo ↗
R / 03 · 2021–2022TEMPORAL FORECASTING

Energy consumption prediction

Built a forecasting approach for electricity demand in Da Nang, Vietnam, covering three-day, one-week and one-month horizons. Used LSTM and Transformer-based time-series models and synthetic data to support model training.

LSTMTransformerTime series
See industry experience ↗
EDUCATION & ACADEMIC RECOGNITION

A continuous learning path.

Formal study, funded scholarships, research and technical competitions, from core computer science through advanced AI and statistics.

University of Cambridge

DOCTORAL STUDIES · PART-TIME
PhD in Mathematics & Statistics
100% funded scholarship (CV)
  • Continuing doctoral study in mathematics and statistics alongside professional work.
  • Research interests connect statistical modelling, machine learning, quantitative finance and model risk.
  • Builds upon previous AI research and practical model development in regulated financial services.
MathematicsStatisticsPhD

University of Surrey

POSTGRADUATE · DISTINCTION
MSc in Artificial Intelligence
80.25 / 100 · Top 1% (CV)
  • Dissertation in deep survival analysis; research accepted at ML4H 2023.
  • Core study: deep learning, applied machine learning, NLP, computer vision, mathematics and statistics, AI & law.
  • Vice-Chancellor’s Excellence Scholarship (2023).
  • Second prize in the 2023 GenAI competition and publication acceptance.
ML4HDeep learningAI research

Troy State University

UNDERGRADUATE · DISTINCTION
BSc in Computer Science
GPA 3.68 / 4.00 · Fully funded
  • Studied probability, statistics, calculus, advanced mathematics, algorithms and object-oriented programming.
  • Microsoft Imagine Cup — first prize (as listed in CV).
  • Intel-sponsored national science and technology competition — second prize.
  • Volunteered at APEC Economic Leaders’ Week and P2A–ASEAN in One.
Computer scienceAlgorithmsScholarship

Academic degrees, achievements and competition results are presented as supplied in the 2026 CV. Publication and demo links point to external research resources.

05 / Tools & methods

The toolkit.

From modelling research to production engineering and risk governance.

01 / Machine learning & research

PyTorchTensorFlowTransformersNLP & LLMsComputer VisionAnomaly DetectionTime Series

02 / Financial risk & quantitative

Credit risk modellingPD / LGD / EADValue at RiskMonte CarloIFRS 9BaselModel validation

03 / Engineering & data

PythonSQLSASSparkAWSAirflowMLflowDockerMinIO

04 / Enterprise platforms

ThetaRayNetRevealIAEPowerCurve Strategy ManagerSageMakerPower BIHUEAthena
06 / LET'S BUILD SOMETHING MEANINGFUL

Better models.
Better decisions.

Interested in financial AI, quantitative modelling, responsible machine learning or research collaboration? I'd be glad to connect.

© 2026 HUNG LE NHAT   /   BUILT FOR CLARITY AND IMPACT¹ Santander financial crime; ² Santander credit risk; ³ Barclays internship — CV-reported figures, not live metrics. Company logos identify past and present employers; this is an independent personal portfolio.BACK TO TOP ↑
Privacy & visit information

If you choose Allow, the portfolio owner receives one Telegram message when you visit, containing the approximate city, region and country inferred from your network IP address, the date/time, page path, referring website domain (if available) and device category. This does not identify you personally and can be inaccurate, particularly when using VPNs or mobile networks. The site does not store visitor records or send raw IP addresses to Telegram. Vercel handles the connection and derives approximate geography; Telegram processes the notification and retains messages under its own policies. You can decline without affecting the site. Your choice is remembered only in this browser tab. Use the Privacy choices control below to change it for future visits; an alert already sent cannot be recalled. For questions, contact the portfolio owner using the email link above.