I'm a Data Platform Architect with 8 years of experience building lakehouse architectures, semantic dimensional models, and AI applications that solve real problems. I've led platform modernizations that reduced development cycles from weeks to days, contributed to 99.9% SLA reliability at scale, and built production RAG systems with LangChain.
My work is guided by a simple principle: technology should improve people's lives. Whether designing data platforms for self-service analytics, building RAG systems for information access, or creating tools for elder care and music education, I focus on practical solutions that create measurable value.
Data Architecture & Platform Leadership
- Designed medallion pattern lakehouses with semantic dimensional models that reduced dashboard development from weeks to days
- Led infrastructure modernization from manual operations to IaC with orchestration pipelines, automated testing, and data quality frameworks
- Built reusable dimensional models enabling self-service analytics across finance, product, and sales teams
AI & Production Systems
- Built production RAG pipelines with LangChain, multiple vector databases (ChromaDB, Pinecone, pgvector, Astra DB), and local LLM integration
- Contributed to 99.9% SLA reliability for platforms serving millions of users through SRE practices and observability work
Leadership & Collaboration
- Led development teams and mentored engineers with a focus on code quality, documentation, and knowledge sharing
- Built developer experience tooling, including CLI tools, unified development environments, and automated quality gates
Data Architecture & AI: Building production-grade data platforms and AI applications. Recent work includes comprehensive lakehouse implementations with medallion architecture, semantic dimensional models for self-service analytics, and RAG systems with LangChain. Experienced with modern orchestration and transformation tools, actively developing expertise in cloud-native lakehouse platforms.
Open Source & Community: Contributing to projects that democratize access to technology and knowledge, including AI agent tooling, language learning applications, and platforms for elder care and music education.
Production RAG System | Python, LangChain, ChromaDB, Vector Databases
Complete RAG pipeline for Finnish immigration information, built with production engineering practices (comprehensive tests, type checking, automated quality gates). Demonstrates end-to-end AI application development from web crawling through semantic search to an interactive chatbot.
Technical Details
Architecture & Implementation:
- Modular architecture separating web crawling, content parsing, vectorization, semantic search, and chat interface
- LangChain framework with text splitting strategies, HuggingFace embeddings, and chain orchestration
- Multi-database support with abstraction layer (ChromaDB, Pinecone, pgvector, Astra DB)
- Local LLM integration with Ollama for privacy-focused inference
- Interactive chatbot interface with Gradio
Engineering Practices:
- Comprehensive test suite with pytest
- Type checking with mypy
- Automated quality gates with ruff and pre-commit hooks
- Modern Python tooling with uv
Lightweight Format for AI Agent Capabilities | Markdown, YAML, AI Frameworks
Collection of skills in the Agent Skills format, an open format for extending AI agent capabilities with specialized knowledge and workflows. Demonstrates practical approaches to making AI agents more capable and context-aware.
AI-Powered Language Learning | Python, AI/ML Frameworks
Interactive web application combining AI with structured lessons to create immersive, context-specific language learning experiences for realistic scenario practice.
Technical Development Tool | Flutter, Dart, MIDI
Mobile application for piano students and teachers focusing on technical development through interactive exercises, real-time feedback, and progress tracking. Built with precise MIDI integration to complement repertoire practice with targeted exercises for coordination and muscle memory.
Data Platform Engineering
Python (Advanced) • SQL (Advanced) • Apache Airflow • dbt • Dimensional Modeling • Semantic Layers • Data Quality Frameworks
AI & ML
LangChain • RAG Architecture • ChromaDB • Pinecone • pgvector • HuggingFace • Ollama • Semantic Search • Vector Embeddings
Cloud & Infrastructure
GCP • AWS • Azure (developing) • Terraform • Docker • Kubernetes • PostgreSQL • BigQuery • Infrastructure as Code
Web & App Development
FastAPI • Django • Svelte • Vue.js • Flutter • Gradio
View Detailed Skills Breakdown
- Data Platform Design (Lakehouse, Warehouse, Medallion patterns)
- Dimensional Modeling (Star/Snowflake schemas)
- Semantic Layer Design
- Data Quality and Governance frameworks
- API Architecture and Management
- Orchestration: Apache Airflow (Advanced)
- Transformation: dbt (Advanced)
- Storage: BigQuery, Data Lakes, Object Storage
- Visualization: Looker/LookML, Looker Studio, Tableau, Metabase
- GCP: BigQuery, Cloud Storage, Cloud Composer, Terraform (Intermediate)
- AWS: S3, EC2, Lambda, RDS (Intermediate)
- Azure: Actively developing expertise
- RAG Implementation and Semantic Search architectures
- LangChain Framework (text splitting, embeddings, chain orchestration)
- Vector Databases (ChromaDB, Pinecone, pgvector, Astra DB)
- LLM Integration and Prompt Engineering
- Statistical Modeling
- Modern Python tooling (uv, ruff, mypy, prek, pytest)
- API frameworks (FastAPI, Flask)
- Web UIs (Gradio, Vue, Svelte)
- JavaScript/TypeScript
- Container Orchestration (Docker, Kubernetes)
- CI/CD and DevOps practices
- English (Native)
- Finnish (Conversational, 10+ years in Finland)
Elder Care Ecosystem | Vue.js, Meteor.js, Django, MongoDB, Plotly.js
Comprehensive toolkit for elder care that makes quality-of-life activities visible and measurable, built from real-world insights gathered through volunteer work at care facilities in Tampere, Finland.
Components:
- Wellbeing: Activity logging and visualization for resident engagement tracking
- Caregiving: Insights platform for coordinators across multiple care levels
- Companionship: Tools for promoting meaningful social connections
Personal Fitness Tracker | Svelte, SvelteKit, TailwindCSS
Open-source workout tracker that helps users visualize progress and maintain consistent exercise habits through simple, effective logging.
I'm actively developing expertise in:
- Cloud-Native Lakehouse Platforms: Exploring Databricks, Microsoft Fabric, and Snowflake architectures
- Advanced Dimensional Modeling: Deepening expertise in semantic layer patterns and metric frameworks
- AI/ML Engineering: Expanding RAG patterns, LLM fine-tuning, and production ML deployment practices
- Data Quality & Governance: Advanced testing frameworks and governance patterns for enterprise data platforms
I believe effective technology starts with understanding real needs. My approach combines:
- Architectural Thinking: Focus on patterns and principles that transfer across platforms rather than vendor-specific implementations
- Technical Excellence: Production-grade engineering with testing, type checking, and maintainable code
- Practical Solutions: Tools that solve actual problems, not technology for its own sake
- Knowledge Sharing: Strong documentation, mentoring, and building systems that help others succeed
- Continuous Learning: Honest about what I know and what I'm developing, with demonstrated ability to learn quickly
Beyond professional work, I volunteer as an English teacher and pianist at Mummon Kammari elder care facility in Tampere, serve as a steward for Tampere Friends (Quakers), and create Creative Commons-licensed music available at brylie.music.
I'm interested in building data platforms and AI applications that create a meaningful impact. If you're working on problems in healthcare, education, sustainability, or other areas with real social value, I'd be glad to connect.
- 📫 Email: brylie@protonmail.com
- 🌐 Website: brylie.online
- 🎵 Music: brylie.music
- 🐦 Bluesky: brylie.bsky.social
- 👔 LinkedIn: brylie






