Miguel Sánchez Duránsenior software engineer · AI systems · DubaiSoftware engineering and applied AI.
working onapplied ai, data analysis, machine learning, full-stack software, agent workflows
I combine full-stack engineering with applied AI, data analysis, and model evaluation.
Work and projects
Explore the workResearch
At work, I help build research tools for finding information in documents and checking answers against sources.
AI experiments
Small experiments with models and evaluation, including the approaches that don’t work out.
Developer tools
Things I build to make day-to-day coding easier, from reviewing changes to following agent sessions.
Fifteen years. Recent work first.
Weighted to current AI and full-stack work; earlier roles in brief.
Senior Software Engineer, AI / full-stack
- Working on retrieval, information extraction, and citation checks across multiple data sources.
- LLM orchestration, LoRA/QLoRA fine-tuning, and evaluation on held-out datasets.
Senior JavaScript Engineer
- Full-stack platforms for government, healthtech, and e-commerce in regulated, sensitive-data environments.
- Early NLP, retrieval pipelines, and LLM-assisted features as the tooling matured.
Earlier engineering roles
Education
MIT
Applied AI and Data Science Program
2026–2027
Universidad de Sevilla
MSc, Web & Mobile Application Development
2011–2012
BSc, Computer Software Engineering
2007–2011
Notes from building careful AI.
What 15 Years of Software Engineering Taught Me About AI Engineering
Which classical engineering disciplines survive contact with stochastic systems, and which problems are genuinely new.
Building Reliable Agentic AI Systems
The Silent Zero: Proving Absence in Very Large Data
How I work
Full-stack development
Connecting interfaces, services, and data.
Evaluation
Testing models and checking answers against source records.
Sensitive data
Local models and access controls where the work calls for them.
Coding with agents
Using agents with tests and review around their changes.
Selected tools
- AI & NLP
- RAG · LangGraph · MCP · spaCy · Sentence Transformers
- Model training
- PyTorch · Hugging Face Transformers · PEFT · LoRA / QLoRA · bitsandbytes
- Data & evaluation
- Python · Pandas · NumPy · Time series · Cross-validation
- Frontend
- Angular · TypeScript · React · RxJS · NgRx
- Backend & databases
- FastAPI · Node.js · GraphQL · MongoDB · SQL
- Delivery
- AWS · Azure · Docker · CI/CD · TDD
Let's build something real.
An engineering problem, a role, or a question about one of the tools: any of those is a good reason to write.