Anthony Roger

I teach software engineeringas a coherent whole— from functional requirements to deployment, without breaks.

The central idea: a good solution arises from a continuous chain between clarity of needs, architectural decisions, and technical rigor. No silos.

What I teach concretely:

- Fundamentals grounded in meaning: semantic analysis of structures, invariants and preconditions, then TDD

- Iterative development driven by specifications

- End-to-end consistency: from business needs to delivered code

- Real delivery: CI/CD, containerization, production packaging

Architect within a transverse architecture team, supporting a 20-year-old legacy medical ERP used by 1000+ hospitals, without direct authority over 27 teams.

- Design and promotion of a model for integrating architects within teams, impact-oriented.

- Proposal for a systematic pairing system between architects to mitigate cognitive stress and maintain knowledge.

- Structuring of a consulting process for management — moving to documented proposals with multiple scenarios, improving decision clarity and reducing political ambiguity.

- Structured decomposition of complex transverse problems into independent root causes, reducing cognitive load and improving the communication of trade-offs.

- Design and implementation of a bidirectional traceability system for requirements compliant with European MDR (2017/745) across 27 teams.

- Platform modernization (Java 6→8, build system restoration, SaaS/OpenShift migration) without disrupting the operations of the 27 teams.

Took charge of a legacy trading platform (Chicago Mercantile Exchange + other markets) managing real-time order orchestration with latency constraints of <150ms. Rebuilt delivery capability, code clarity, and client trust.

- Identification of latency bottlenecks in the client workflow. Proposal and sale of optimization (separate invoices), then implementation and delivery.

- Reduction of release cycle from 2 days to 2 hours through end-to-end delivery chain automation.

- Maintenance of <150ms latency SLA in production.

- Stabilization of project support to less than 1 incident per week.

- Re-specification of core functionalities (order management, instrument aggregation) restoring clarity and maintainability of undocumented codebase.

- Restoration of test suites across 3 business domains (~100 tests each): price aggregation, price distribution, order orchestration.

- Design and implementation of behavior-driven black-box integration tests, written before implementation using a custom framework.

- Implementations with full traceability for audit compliance and incident resolution.

- Arbitration between business, architecture, performance, and management stakeholders to define a balanced technical and delivery roadmap.