ML Developer

About the Role

We are looking for an ML Developer with 6+ years of experience to design, build, and deploy machine learning solutions supporting the surveillance function within a financial services environment. You will apply advanced ML techniques — anomaly detection, NLP, behavioural analytics, and classification models — to enhance communication and trade surveillance capabilities, helping detect market abuse, misconduct, and compliance breaches at scale.

Key Responsibilities

• Design, develop, and deploy ML models for surveillance use cases — anomaly detection, behavioural analytics, NLP-based communication analysis, and pattern recognition across large-scale financial data.
• Build and maintain NLP pipelines — text classification, sentiment analysis, entity recognition, and semantic similarity — for electronic communication surveillance.
• Develop and maintain feature engineering pipelines — extracting, transforming, and selecting relevant features from trade, communication, and market data for ML model training.
• Train, evaluate, and optimise ML models — applying supervised, unsupervised, and semi-supervised learning techniques for surveillance detection and alert prioritisation.
• Integrate ML models into surveillance platforms — deploying models as REST APIs or batch inference pipelines for production consumption.
• Implement model monitoring and drift detection — ensuring surveillance models maintain accuracy, reliability, and performance in production.
• Collaborate with surveillance analysts, compliance teams, and data engineers — translating surveillance business requirements into well-designed ML solutions.
• Support alert tuning and false positive reduction — applying ML-driven approaches to improve surveillance detection quality and operational efficiency.
• Produce clear technical documentation — model design documents, feature specifications, and deployment runbooks.

What We're Looking For

• 6+ years of ML development experience with strong Python and machine learning expertise in production environments.
• Strong ML skills — supervised, unsupervised, and semi-supervised learning — applied to anomaly detection, classification, and behavioural analytics use cases.
• Proven NLP expertise — text classification, sentiment analysis, named entity recognition, and semantic search using spaCy, NLTK, Hugging Face, or equivalent.
• Strong Python proficiency — NumPy, pandas, Scikit-learn, TensorFlow, or PyTorch — for ML model development and data processing.
• Experience with feature engineering — extracting and transforming features from structured and unstructured financial data for ML model training.
• Hands-on ML model deployment experience — REST API serving, batch inference, and containerised model deployment using Docker.
• Experience with MLflow or equivalent — experiment tracking, model versioning, and model registry management.
• Solid SQL skills — data querying and validation across large-scale financial datasets.
• Good understanding of financial services compliance or surveillance domain — communication surveillance, trade surveillance, or market abuse detection.
• Strong AWS experience — SageMaker, S3, Lambda, or equivalent — for ML model training and deployment.
• Excellent communication skills — presenting ML findings, model performance, and surveillance insights to compliance and technology stakeholders.

Nice to Have

• Experience with LLMs and GenAI — GPT, Claude, or Llama — for advanced communication surveillance and intelligent alert summarisation.
• Familiarity with surveillance platforms — NICE Actimize, Behavox, or equivalent — and their ML integration patterns.
• Knowledge of graph-based ML — network analysis and entity relationship modelling for market abuse detection.
• Exposure to real-time streaming ML — Kafka or AWS Kinesis — for real-time surveillance alert generation.
• Experience with explainability frameworks — SHAP, LIME — for model transparency and regulatory audit support.
• Knowledge of regulatory frameworks — MiFID II, MAR, ASIC Market Integrity Rules — and their surveillance implications.
• AWS Certified Machine Learning Specialty certification.
• Experience in Financial Services, Banking, or regulated industry environments.