Data Engineer and Full Stack Developer with 8+ years building production systems across fintech, analytics, and AI. Currently based in Douala, Cameroon, Nigeria and France.

• 🔭 Currently: Data Engineer at Ejara — building real-time financial data infrastructure for a crypto/fintech platform using Spark, Kafka, Iceberg, and Debezium CDC
• 🏗 What I ship: Multi-module streaming pipelines, full-stack analytics products, and ML systems — end to end
• 🌱 Deepening expertise in: Distributed systems, LLM applications, and real-time data architectures
• 💬 Ask me about: Data streaming, CDC pipelines, Python, Java, FastAPI, React, or Computer Vision
• ⚡ Background: Started in Computer Vision, moved into Data Engineering — now I combine both

Data Engineer @ Ejara — Crypto/Fintech Platform (2023 - Present)

• Built and maintain a multi-module real-time streaming pipeline (Apache Spark + Kafka + Iceberg + Polaris) processing financial data from 5 independent sources into a unified data lake
• Implemented Debezium CDC capturing changes from PostgreSQL, MySQL, and MongoDB into Iceberg in real time
• Streaming sources: crypto market data (296 pairs), Amplitude analytics, Elasticsearch logs, ECB exchange rates (30 currencies × 5 years of history), and database CDC
• Manages the full data platform: Airflow batch pipelines, Snowflake transforms, schema registry (Apicurio), and S3/Polaris catalog
• Owns infrastructure: Docker, watchdog automation, zero-downtime deploys via custom shell tooling

Production real-time data platform for a crypto/fintech company. Multi-module Maven project (Java + Spark + Kafka + Debezium + Apache Iceberg) with 5 independent streaming pipelines: CDC from relational databases, crypto market data (296 pairs), Amplitude analytics, Elasticsearch logs, and ECB exchange rates. Custom watchdog, zero-downtime deploy tooling, and Polaris catalog for Iceberg metadata. Full-stack analytics platform — REST API backend + React client. Built for business intelligence and data-driven decision making. Full-stack product with core API, frontend, and a dedicated security layer. Production system with authentication, authorization, and data management. ML-based customer segmentation using Recency, Frequency, and Monetary analysis. End-to-end pipeline from raw transaction data to actionable customer clusters.

Microservice-based e-learning platform with API, payment integration, booking system, and centralized database. Migrated from monolith to microservices architecture.

Robust data extraction pipeline for complex document structures using Mistral and Claude. Handles unstructured documents at scale.

• Designed and shipped a real-time streaming platform from scratch
• Built full-stack independently — API, frontend, and infrastructure
• Transitioned a monolithic e-learning platform to a microservices architecture
• Background in Computer Vision (PyTorch/TensorFlow) applied to document extraction and business prediction systems

• 📧 Email: tafowocedric@gmail.com
• 💼 LinkedIn: Tafowo Cedric
• 📱 WhatsApp: +237 680752220
• 💬 Telegram: @tafowocedric

"Ship systems that work. Understand data that matters."