Senior Data Architect | Full Stack AI Engineer | MLOps/LLMOps | Data Scientist

I am a hands-on architect and developer based in Toledo, Spain, currently working as a remote Data Architect. With over a decade of experience bridging the gap between massive, distributed data warehouses and interactive discovery environments, I like to architect systems that treat performance as a feature, not an afterthought.

I also build many AI & LLM based solutions for reducing the negative impact of small language models in agentic systems, everything local-first and thinking on production from the first line of code

• AI & Agentic Workflows: Designing complex, multi-agent systems using LangChain and LangGraph (my preferred framework for tight control over sub-agents).
• Local AI & Homelabs: Deploying personal AI inference servers and self-hosting solutions using LiteLLM, Nginx, vLLM, and Docker on VPS and local hardware.
• Data Engineering Environments: Championing lightweight, high-performance, local-first data stacks combining tools like DuckDB, RustFS, Neo4J, PostgreSQL, ClickHouse, Databricks, MongoDB, Redis, and InfluxDB.
• Workflow Orchestration: Building reliable pipelines and automation flows with Airflow and Dagster.
• Community & Gaming: Developing the Sendou.ink SDK for the competitive Splatoon 3 bot usage.

Languages:

Data & Analytics:

AI & Machine Learning:

Cloud & Infrastructure:

• Data Buddy: An AI-powered data engineering assistant for making POCs or learning purposes, made with Langchain, Arize Phoenix, and a lean Django-based background worker. It abstracts the friction of modern data orchestration, helping teams generate ETL scripts, SQL, and Jupyter notebooks securely.
• Sendou.ink Python SDK: A fully Pythonic, developer-friendly interface for the Sendou.ink API. Designed to facilitate data extraction and automation for the competitive Splatoon community.
• OBS-Switch: A Python package enhancing livestreaming production by enabling Nintendo Switch and OBS automations via webhooks and websockets.
• Local Secrets: A Golang made AWS Secrets replica for using it with its own Python client, using AWS CLI or using Boto3 in Python, other AWS clients can be used as soon that it let you change the endpoint.

I frequently publish technical deep-dives on Medium. Some of my focus areas include:

• Data Engineering in a Box: A comprehensive 13-part architectural series on building a complete, localized data lakehouse using Docker, MinIO, PostgreSQL, PySpark, Dremio, and DuckDB.
• Quick Local Sentiment Analysis: Practical, 15-minute guides on building sentiment analysis, predictive maintenance, and fraud detection pipelines using Faker, DuckDB, and Scikit-Learn.
• Algorithms Toolkit: A Python developer's guide translating dense academic algorithms (from Binary Search to $A^*$ and FFT) into practical, everyday engineering tools.
• Real-Time Architectures: Exploring event-driven solutions (MQTT, Kafka, PostGIS) to prevent server overload and device battery drain in massive-scale tracking apps.

• GitHub: @jasonjimnz
• Medium: @jasonjimenezcruz
• LinkedIn /in/jasonjimenezcruz