• Principal Architect — 21 years in tech space (C/C++ → Java/Scala → Python/TypeScript → Prompt/Agent)
• Specializing in context engineering and legacy estate modernization.

Working with Tier-1 European banks and G-SIBs across:

• Regulatory Reporting & Data Lineage (RegTech) Regulators mandate traceable lineage — free-text transformation rules fail BCBS 239 audits, break AI reasoning, and expose banks to material fines. Modernizing Basel III/IV, BCBS 239, FRTB, CRR3, and COREP pipelines with machine-readable lineage contracts that surface hidden dependencies before migration, satisfy Principle 2 automatically, and make every regulatory data asset trustworthy enough for an agent to reason over.

• Analytics Estate Exit (SAS) Migrating SAS workloads — Base SAS, SAS/STAT, DI Studio, Model Manager, Visual Analytics, Viya, and AML suites — to Databricks and Snowflake using automated transpilation and semantic uplift, preserving statistical logic while eliminating platform lock-in.

• EDW Exit (Teradata · Netezza · Oracle · MS SQL) Migrating large-scale EDW estates to Databricks and Snowflake — schema translation, workload profiling, query optimization, and medallion architecture target design that eliminates the performance and cost burden of on-premise warehousing.

• ETL & Integration Modernization (DataStage · Informatica · ODI · Talend) Retiring on-premise ETL platforms and re-engineering pipelines as cloud-native data products — embedded data quality, machine-readable lineage contracts, and observable orchestration replacing opaque batch jobs.

• Mainframe & Batch Orchestration Exit (COBOL · JCL · AutoSys · Control-M · OPC · UC4) Modernizing mainframe-bound workloads and legacy job schedulers into event-driven, cloud-native pipeline orchestration — extracting decades of embedded business logic into testable, versioned, agent-consumable services while preserving SLA contracts and gaining full observability.

I work at the boundary between technical architecture and commercial decision-making — helping enterprise buyers move from problem recognition to funded programme.

• Context Engineering Designing context supply chains for enterprise AI — semantic models, data contracts, knowledge graphs, and retrieval pipelines that give agents accurate, governed, auditable answers.

• Legacy Estate Modernization Architecture, scoping, and automated discovery across a broad range of on-premise platforms — mainframe, SAS, legacy EDW and ETL — into modern cloud-native data products.

Databricks Snowflake GCP Python SQL
Basel III/IV BCBS 239 FRTB Medallion Architecture Data Mesh Knowledge Graphs

• Semantic layer as the missing link — ontologies, OWL/RDF models, and machine-readable contracts that turn migrated estates into assets an agent can actually trust and reason over, without platform lock-in
• Model Context Protocol (MCP) as an emerging enterprise standard for connecting agents to governed data sources at scale
• GraphRAG — knowledge graph-augmented retrieval for regulatory and compliance domains where relationship traversal matters as much as similarity search
• Agentic RAG vs. long-context tradeoffs — when retrieval pipelines outperform brute-force context stuffing in regulated, high-stakes domains
• Context evaluation frameworks — measuring context quality, coverage gaps, and hallucination risk before agents reach production
• Agentic memory architectures — designing episodic and semantic memory layers that let enterprise agents learn from feedback without retraining

• Agentic architecture and Context Engineering assessments
• BFSI data estate modernization advisory
• Agentic AI and legacy estate migration

📍 London, UK  |  LinkedIn