Snowflake, dbt, FastAPI, Next.js — from data platform architecture to production-ready AI products. Built for reliability, shipped with speed.
Production builds across data, AI, SaaS, security, and Web3 — each with a live link or case study.
Autonomous quant trading platform with AI-driven strategy selection, multi-agent consensus, Hyperliquid SDK execution, and live sentiment analysis.
Zero-knowledge encrypted document vault with client-side XChaCha20-Poly1305 encryption, SDK-first architecture, and GDPR Article 17 data sovereignty.
AI compliance operating system for NIS2 and DORA workflows.
Open-source enterprise data platform — catalog, column-level lineage, pipeline monitoring, FinOps, secrets management, and governance with 14+ scanner integrations.
Modular DeFi product spanning analytics, liquidity, governance, and community play.
Developer email toolkit — syntax-highlighted snippets, encrypted token sharing, and secure file delivery from your inbox.
Autonomous offensive security platform — 39 AI exploit agents, 25+ scanners, multi-LLM orchestration, and DORA TLPT compliance in a single pane of glass.
B2B growth platform with analytics-led workflows and premium operational UX.
AI workflow platform tailored for the French market with production-ready delivery patterns.
Autonomous quant trading platform with AI-driven strategy selection, multi-agent consensus, Hyperliquid SDK execution, and live sentiment analysis.
Zero-knowledge encrypted document vault with client-side XChaCha20-Poly1305 encryption, SDK-first architecture, and GDPR Article 17 data sovereignty.
AI compliance operating system for NIS2 and DORA workflows.
Open-source enterprise data platform — catalog, column-level lineage, pipeline monitoring, FinOps, secrets management, and governance with 14+ scanner integrations.
Modular DeFi product spanning analytics, liquidity, governance, and community play.
Developer email toolkit — syntax-highlighted snippets, encrypted token sharing, and secure file delivery from your inbox.
Autonomous offensive security platform — 39 AI exploit agents, 25+ scanners, multi-LLM orchestration, and DORA TLPT compliance in a single pane of glass.
B2B growth platform with analytics-led workflows and premium operational UX.
AI workflow platform tailored for the French market with production-ready delivery patterns.
Flagship case studies and selective product snapshots focused on decisions, proof, and delivery quality.
Autonomous quant trading platform with AI-driven strategy selection, multi-agent consensus, Hyperliquid SDK execution, and live sentiment analysis.
Zero-knowledge encrypted document vault with client-side XChaCha20-Poly1305 encryption, SDK-first architecture, and GDPR Article 17 data sovereignty.
AI compliance operating system for NIS2 and DORA workflows.
Open-source enterprise data platform — catalog, column-level lineage, pipeline monitoring, FinOps, secrets management, and governance with 14+ scanner integrations.
Modular DeFi product spanning analytics, liquidity, governance, and community play.
Developer email toolkit — syntax-highlighted snippets, encrypted token sharing, and secure file delivery from your inbox.
Autonomous offensive security platform — 39 AI exploit agents, 25+ scanners, multi-LLM orchestration, and DORA TLPT compliance in a single pane of glass.
B2B growth platform with analytics-led workflows and premium operational UX.
AI workflow platform tailored for the French market with production-ready delivery patterns.
In data: analyst, manager, PO, architect
Shipped to production
Pernod Ricard, Air Liquide
Reliability, governance, speed
I am Cedric, Southern Europe Lead Data Architect with 8 years of experience across data analysis, product ownership, data management, and architecture — building high-performing data products and modern AI-enabled solutions for enterprise teams.
My core stack includes Snowflake, dbt, Matillion, FastAPI, Next.js, and Python automation. I focus on reliability, performance, and pragmatic delivery.
I translate business requirements into technical designs, ensure data quality, and drive architecture decisions across ingestion, transformation, and consumption layers.
Daily operations include ETL monitoring, incident handling, and end-to-end ownership for critical data refreshes.
I build and iterate quickly, then publish only the parts that are safe to share publicly: the framing, the constraints, the decisions, and the outcomes.
This portfolio is designed to show judgment and delivery quality without exposing sensitive implementation details.
I partner with founders and teams from framing to rollout: roadmap, architecture, implementation, and stabilization.
You get clear communication, fast iteration loops, and resilient technical foundations.
End-to-end data and AI execution — every tool, framework, and pattern I've used across 9 production products.
End-to-end design of scalable data foundations: ingestion pipelines, medallion layers, modeling standards, orchestration, and governed consumption.
Warehousing & Compute
Snowflake
Primary cloud DWH — multi-cluster, time travel, zero-copy clones
PostgreSQL
Operational DB for apps, audit logs, metadata stores
Supabase
Managed Postgres with auth, realtime, and edge functions
Redis
Cache layer, session store, job queue broker
Ingestion & Transformation
dbt
SQL-based transformation layer with testing and docs
Matillion
ELT orchestration for Snowflake at enterprise scale
Azure Data Factory
Cloud ETL pipelines and data movement
SQLAlchemy + Alembic
Python ORM with migration management
Orchestration & Storage
ARQ
Async Python task queue for background jobs
Celery + Beat
Distributed task queue with periodic scheduling
MinIO
S3-compatible object storage for encrypted blobs
Docker Compose
Multi-container orchestration for local + staging
Projects
Design and ship AI-enabled workflows with RAG pipelines, agent orchestration, observable behavior, and production guardrails.
Frameworks & Agents
LangChain
Agent orchestration, chains, tool calling
PydanticAI
Type-safe AI workflows with structured outputs
FastAPI
Async Python API layer powering all AI backends
OpenRouter
Multi-model routing with fallback strategies
Models & Embeddings
Mistral AI
French-first LLM for compliance and generation
Ollama
Local model inference for dev and testing
Sentence Transformers
Embedding models for semantic search
Qdrant
Vector database for similarity search
Observability
LangFuse
LLM tracing, evaluation, and cost analytics
OpenTelemetry
Distributed tracing across AI services
Grafana
Dashboards for backend and AI metrics
Prometheus
Metric collection and alerting
Projects
Frontend and backend implementation for production SaaS products — from component architecture to API design, auth, billing, and deployment.
Frontend
Next.js 14/15
App Router, RSC, SSR, ISR, edge runtime
React 18
Component architecture, hooks, Suspense
TypeScript
Type safety across every frontend and SDK
Tailwind CSS
Utility-first styling with design tokens
State & Data Fetching
TanStack Query
Cache management, background refresh, optimistic updates
Zustand
Lightweight client state for complex UIs
React Hook Form + Zod
Type-safe form validation
Axios
HTTP client with interceptors and retry logic
Backend & Services
Symfony 6
PHP backend with Doctrine ORM and Twig
Prisma
Type-safe ORM for Node.js backends
Stripe
Subscription billing and payment processing
Clerk
Auth provider with SSO and session management
UI & Visualization
Framer Motion
Animation library for React interfaces
Three.js + R3F
3D visualization for landing pages
Chart.js
Data visualization and analytics charts
CodeMirror 6
Code editor with syntax highlighting
Projects
Zero-knowledge architectures, client-side encryption, tamper-proof audit logs, and secure sharing protocols built for real compliance.
Encryption & Keys
XChaCha20-Poly1305
AEAD cipher with 192-bit nonces — no reuse risk
Argon2id
Password-based key derivation (OWASP params)
NaCl SecretBox
Symmetric DEK wrapping for per-document keys
NaCl Sealed Box
Asymmetric re-encryption for secure sharing
Libraries
libsodium (WASM)
Client-side crypto running in the browser
PyNaCl + cryptography
Server-side crypto utilities in Python
Libsodium (PHP)
Token encryption at rest in Symfony backend
python-jose
JWT authentication and token management
Compliance & Audit
HMAC-SHA256 Chains
Tamper-proof audit log with chain integrity
GDPR Erasure Protocol
Cryptographic deletion workflows
Forensic Watermarking
Leak detection on shared documents
Scoped API Keys
Granular access control with revocation
Projects
Operational tooling, background workers, webhook delivery, CI/CD pipelines, and observability to keep production systems healthy.
Workers & Queues
ARQ
Async Python workers for expiry, cleanup, webhooks
Celery + Beat
Distributed task queue with scheduled jobs
Redis Pub/Sub
Event-driven messaging between services
Resend
Transactional email delivery with templates
Infrastructure
Docker
Containerized services across all projects
Nginx
Reverse proxy with TLS termination
Vercel
Edge deployment for Next.js frontends
Caddy
Automatic HTTPS for self-hosted services
Monitoring
Grafana
Operational dashboards and alerting
Prometheus
Metric collection and time-series storage
Sentry
Error tracking and performance monitoring
LangFuse
AI-specific cost and latency tracing
Projects
Strategic support for data roadmaps, architecture reviews, team coaching, and execution governance — from framing through production.
Strategy & Process
Jira + Confluence
Sprint planning, ADRs, technical documentation
Product Roadmapping
Priority frameworks, OKR alignment
Data Mesh Patterns
Domain ownership, data contracts, self-serve
Architecture Reviews
System design, trade-off analysis, RFC process
Delivery & Governance
Quality Frameworks
Data freshness, accuracy, completeness checks
Incident Response
Runbooks, escalation paths, post-mortems
Cost Governance
Warehouse spend tracking and optimization
Team Coaching
Pair programming, code reviews, mentoring
Enterprise Context
Pernod Ricard
Southern Europe data architecture leadership
Air Liquide
CAC 40 enterprise data platform experience
NIS2 & DORA
Regulatory compliance workflows
GDPR
Data sovereignty and privacy engineering
Projects
The same four-phase workflow I've used to ship 9 products — from Comply-Agent and CoffreZero to enterprise data platforms at Pernod Ricard and Air Liquide.
Every product starts with a problem worth solving. I map the user pain points, define success metrics, identify constraints, and scope the first deliverable. Whether it's a compliance gap (Comply-Agent), a trust problem (CoffreZero), or a market inefficiency (Lumanagi) — I frame the product before touching code.
I build a focused proof-of-concept to validate the riskiest assumption first. For AI products, that means a working RAG pipeline with real data. For SaaS, a functional core loop. For security, a crypto prototype that proves zero-knowledge is viable. No throwaway code — the POC is designed to evolve into the MVP.
The MVP ships with auth, core features, and enough polish that real users can give meaningful feedback. I build incrementally with CI/CD, PR reviews, and multi-environment deployment. Frontend, backend, infrastructure, and observability are set up from the start — not bolted on later.
Production means monitoring, hardening, and operational readiness. I set up dashboards, error tracking, security reviews, runbooks, and handover documentation. For data platforms, that includes reconciliation, SLAs, and cost governance. The product runs without me — that's the benchmark.
I work with teams that need strong architecture and fast execution. Share your context and we can define the right delivery model.
Personal AI Assistant
