Bilal Meddah

System design & architecture

• Multi-tenant research workspace on Python/FastAPI microservices: lab and project isolation, fine-grained access control, configurable compute environments.
• Legacy data migration into the new core: schema mapping, re-runnable backfills, validation passes and a staged cut-over by data type.
• Service boundaries, domain models and versioned API contracts for the core entity service, one source of truth across services.

Full-stack product engineering

• Full-stack features across the data, workflows and virtual-lab sections, each shipped from FastAPI endpoint to React screen.
• ~90% of the new React/TypeScript client: server-state caching, normalized stores, virtualized data-heavy views.
• Legacy app migrated to a modular core screen by screen, without a feature freeze; frontend tech debt down by more than 60%.

Quality & reliability

• obi-e2e: 84 scenarios and 300 Playwright cases from plain-English specs, run on preview, staging and production.
• Credential-lifecycle manager: background token refresh keeps long-running jobs authenticated past expiry.
• Run hygiene on every test run: service health preflight, self-provisioned projects, automatic cleanup.

Applied AI

• AI research assistant, built with the AI team: data discovery, summaries, intent-based routing to tools and datasets.
• AI agents that generate and repair Playwright tests from scenarios, reviewed through pull requests.
• Image pipelines for scientific artifacts: normalization, compression, model-driven synthesis.

System design & architecture

• Distributed simulation platform: parallel job orchestration, real-time multi-user sessions, streamed results.
• API layer between the interactive UI and compute: launch, status and results for long-running runs.

Full-stack product engineering

• Collaborative research tools: shared workspaces and data-driven views for daily research workflows.
• React 18 concurrent rendering for responsive, data-heavy interfaces.

Delivery & developer experience

• Publication sites moved from a CMS to JAMStack: 70% faster pages, lower server load.
• React 16 → 18 and Webpack 4 → Vite 6, upgraded in steps: CI/CD 50% faster, builds 40% faster.

Applied AI

• LLM assistant for natural-language queries over scientific data, from API design to product UI.

Quality & reliability

• E2E testing and CI pipelines, shared test libraries and developer utilities.

Technical leadership

• Tech Lead for a team of 5 engineers: roadmap, design reviews, code-review standards, mentoring.

System design & architecture

• Micro-frontend architecture: an app shell for routing, auth and layout, with product modules built and deployed independently.
• Reusable backend components for authentication, validation, notifications and audit logging, shared by every module.
• Role-based access for clinic staff and administrators, enforced on the API and reflected in the UI.

Full-stack product engineering

• Patient, appointment and billing workflows built from database schema and REST endpoints to the UI.
• Shared UI kit (forms, tables, scheduling and date components) used across all modules.
• Landing page 50% faster with HTTP/CDN caching, code-splitting and lazy loading; conversion up 20%.

Delivery & developer experience

• Monorepo with shared packages and one toolchain; CI builds and tests only what a change affects.
• Independent, versioned module releases; the shell loads new modules without a full redeploy.

Quality & reliability

• Unit and integration tests on shared packages, lint and type gates on every merge.
• Error monitoring and structured logging in production.

Technical leadership

• Tech Lead for a team of 5 engineers: architecture decisions, design and code reviews.
• Planning with product: scoped milestones, regular delivery, clear ownership.

Full-stack product engineering

• React Native and Expo apps for iOS and Android from one TypeScript codebase.
• Shared component library and typed API clients across client apps, including Juice Plus+, Kudos and Fimimus.

Delivery & developer experience

• One build pipeline producing releases for both app stores.

Quality & reliability

• Large-scale refactor into feature modules, with documentation next to the code: less maintenance, faster onboarding.