Platform Architecture · Quantitative Systems · Regulated Infrastructure
Lead Quantitative Developer at Investec. 15+ years shipping production systems at scale—from regulated financial infrastructure ($1B+ portfolio) to automotive powertrains. I build elegant platform abstractions that translate complex quantitative and regulatory requirements into self-service tools. Zero audit findings across Big Four reviews. 8x throughput acceleration on critical workflows.
Currently architecting model lifecycle platforms on Azure (Container Apps, Functions, Cosmos DB). Python SDK design & developer experience focused. Patterns: API-led architecture, TDD-first delivery, hierarchical abstraction layers, governed AI-assisted engineering.
Model Lifecycle Platforms
End-to-end platforms spanning development, testing, validation, monitoring, and governance. Audit-grade data lineage, evidence tracking, regulatory frameworks. Pattern: hierarchical abstraction layers enabling distributed model teams to scale independently.
Cloud-Native Architecture
Azure-first systems design: Container Apps for distributed workloads, Function Apps for event-driven execution, polyglot data layer (SQL/Cosmos DB). Observability-first with Application Insights instrumentation and governance-driven RBAC. 99.95% uptime SLA track record.
Python Platform SDKs
Multi-tier SDK strategy: thin client libraries wrapping Azure/internal APIs; domain-specific abstractions for risk models, data pipelines, workflows. Monorepo structure enabling atomic versioning across 15+ consumer teams. Focus on developer experience and API clarity.
Quantitative Systems
Monte Carlo simulation engines, stochastic optimization for capital efficiency, PD/LGD calibration frameworks, time-series forecasting. Numerical methods (gradient descent, multi-objective genetic algorithms). Integration with regulatory capital models (IFRS 9, IRB, FCA frameworks).
| Systems | Python (FastAPI, Pydantic, SQLAlchemy, NumPy) · SQL (T-SQL, optimization, stored procedures) · Bash |
|---|---|
| Cloud & Infrastructure | Azure: Container Apps, Function Apps, SQL, Cosmos DB, Machine Learning, Monitor · Azure DevOps pipelines · Docker |
| DevOps & Reliability | PyTest & TDD, code coverage gates, vulnerability scanning, Azure DevOps CI/CD, Infrastructure as Code (Bicep/Terraform roadmap) |
| Data & Simulation | ETL pipelines, time-series analysis, heterogeneous data ingestion, Monte Carlo simulation, stochastic optimization |
| SDK & Developer Experience | Monorepo architecture, hierarchical client libraries, API design, package management, developer tooling |
Engineering Leadership · Building high-performing teams through regulatory intensity · Platform strategy and technical direction
Model Governance · IFRS 9, IRB, FCA MCOB frameworks · Audit-grade data lineage and evidence tracking
Python SDK Design · Developer experience optimization · Multi-tier architecture patterns · Monorepo governance
AI-Assisted Engineering · Structured context patterns · Reusable agent workflows · Code generation & review best practices
Observability & Reliability · Data pipeline monitoring · Cost-optimized cloud architecture · 99.95%+ SLA design
Investec — Credit Risk Model Platform
Led end-to-end delivery of $1B+ portfolio model estate (IFRS 9, IRB, FCA MCOB 11.6) on Azure cloud-native infrastructure. Zero audit findings across Big Four annual reviews. Built Credit Modelling Portal (100+ concurrent users), Model Development Environment (hierarchical SDKs, 15+ modeller teams), and impact assessment system (2–3 → 8 runs/day throughput acceleration). Delivered Private Client Calculation Engine for HNW mortgage affordability ($500M+ portfolio impact).
Ford Motor Company — Analytics Platform
Led team of 7 analysts as SME for predictive analytics. Automated 70% of recurring workload with SQL/Python tooling. Designed data-processing pipelines for heterogeneous engineering datasets (50M+ daily observations). Enabled large-scale CO₂/NVH predictive modeling across multinational OEM. Introduced GitHub source control and improved team development discipline across 10+ MATLAB collaborators.
Prodrive Automotive — CREO Research Program
Lead engineer on UK Government-funded TSB project (3 universities, 4 industry partners). Developed MATLAB-based predictive emissions models using Neural Networks and Multi-Objective Genetic Algorithms. Simulation methodology adopted as Ford Motor Company standard across commercial vehicle emissions program.
Platform Leadership
Line manager accountable for model team delivery in high-pressure regulatory environments. Strong team retention track record. Selected for Investec's competitive Leadership Academy Development Team Leader programme.
Cross-Functional Alignment
Bridge between quantitative research, IT infrastructure, Risk governance, and executive strategy. Primary technical contact for Big Four audits. Trusted translator of complex quantitative concepts to regulators, business stakeholders, and executive teams.
Execution at Scale
Owned simultaneous delivery of IFRS 9 Point-in-Time redevelopment, LGD framework remodel, impact assessment infrastructure, and ML-based model monitoring—all under active regulatory scrutiny.