About me

• Backend-focused full-stack engineer (Python/Django/DRF + Vue.js) with 15+ years building data-intensive, research-driven applications. Engagements run end-to-end, from requirements through architecture and implementation to production deployment. Domain-agnostic, with a scientific-computing background.
• Scalable Django + PostgreSQL systems with RESTful APIs, caching (Redis/Memcached), message brokers (RabbitMQ) and asynchronous processing (Celery); robust authentication (Authentik/OAuth/JWT) and multi-tenant SaaS (django-tenants).
• Applications containerized with Docker and deployed on Kubernetes, backed by GitLab CI/CD and cloud-native infrastructure. Strict code-quality discipline throughout: type hints, testing, pre-commit hooks, and clear documentation (MkDocs). Agentic coding workflows amplify this: I direct AI-assisted development as the expert human in the loop, and build the tooling itself. Self-hosted LangGraph + Claude Code agents (MCP servers) that carry a tracked issue through plan, self-fixing test loop, merge request, and automated AI review, outbound-only and data-resident.
• Responsive frontends in Vue 3 and Tailwind, tuned for the usability demands of complex workflows. Core developer of the open-source LabID data management and Electronic Lab Notebook platform.

Skills

• My main programming language is Python, while mainly working with the Django framework providing RESTful APIs. I have used R for data analysis and have written Perl and PHP in the past.
• For the frontend, I build interfaces with Vue.js and Tailwind, and create documentation and static sites with generators like Jekyll, MkDocs or Hugo (e.g., the Hugoplate theme). Solid grounding in vanilla JavaScript, jQuery, and CSS(3).
• Throughout my career, I have been dealing with compute servers/clusters to do analysis on, or I connected services like Galaxy to provide these resources to non-technical users. I have set up bare-metal machines and VMs with different Linux or Windows flavors.
• For CI/CD purposes and beyond, I have built Docker (as well as Singularity) images and deployed them with Docker Compose, to a Kubernetes cluster, or on AWS cloud.
• On the backend, I work with PostgreSQL, multi-tenant SaaS (django-tenants), authentication via Authentik/OAuth/JWT, and asynchronous processing with Celery, Redis, and RabbitMQ. For object storage, I use S3/MinIO.
• For AI-assisted development, I build and run LLM agents with LangGraph and Claude Code (headless), exposing tools through MCP servers, provider-agnostic across Anthropic Claude, OpenAI, and AWS Bedrock via LangChain/LangGraph. I've built a self-hosted issue-to-merge-request loop (plan, code, self-fixing test loop, then automated AI review) running as containers on self-hosted GitLab runners (outbound-only, data-resident).
• I am security-conscious and follow security best practices: applying the principle of least privilege (granting only the access that is strictly necessary), secure defaults, careful secret and dependency management, and minimizing the attack surface across the application, deployment, and infrastructure layers.
• In addition to Galaxy, I'm familiar with the Snakemake and Nextflow workflow managers. Relatedly, I have used and written (bio)conda recipes and EasyBuild configs.
• In the past, I have used SVN and Mercurial, but for over 10 years Git has been my go-to software versioning system.