I'm a software engineer with 7+ years of experience designing, building, and maintaining software products and services. My background spans software architecture, backend services, native applications, applied AI, and robotics.

I enjoy solving practical engineering problems, making thoughtful architectural decisions, and building software that is reliable, maintainable, and genuinely useful. Lately, my work has brought together systems programming, AI-powered tools, and real-world experimentation.

• 👨‍💻 Projects & portfolio: danielz.co.uk
• 📝 Technical writing & blog: danielz.co.uk/blog
• 📄 CV: danielz.co.uk/cv
• 📫 Contact: danial.za@outlook.com
• ⚡ Fun fact: I like building AI systems that do real work, not just nice demos.

• Software architecture & product development — building maintainable applications and services from requirements through implementation.
• Backend engineering & integrations — APIs, databases, asynchronous workflows, and the services behind user-facing products.
• Performance & reliability — profiling, benchmarking, resource-aware design, testing, and handling failures thoughtfully.
• Applied AI & developer tools — LLM-powered products, copilots, multi-agent workflows, tool integration, and evaluation.
• Retrieval & data systems — RAG, semantic search, vector databases, and grounded responses.
• Native & cross-platform development — macOS apps, mobile interfaces, and Flutter-based applications.
• Robotics & machine learning — reinforcement learning, control systems, sim-to-real evaluation, and experiments on real hardware.

• Usta — Native macOS Multi-Agent IDE
  A native macOS workspace where specialist AI agents collaborate on software projects. Built with a Swift frontend and a Rust/Tokio backend, using gRPC, terminal sessions, and a shared event bus.

• TalkingHeadAI — Conversational AI
  A conversational AI application combining retrieval, LLM-powered responses, backend APIs, and policy-based response checks, designed around practical interactions and responsive user experiences.

• AEGIS — Robot Safety Research
  Public research data, figures, and reproducibility materials for an age-aware robot safety approach using RGB-D sensing and vision-language monitoring on the Niryo NED3 Pro.

• Adaptive Robot Control — Published Research
  Research code and experimental materials for deploying and evaluating adaptive control on real robot hardware. Published in Sensors (2026).

• Build for people: start with the problem and design around real usage.
• Keep systems understandable: favor clear interfaces, useful abstractions, and maintainable code.
• Measure before optimising: use profiling, benchmarks, and evidence to guide technical decisions.
• Own the whole solution: think about integration, reliability, deployment, and what happens after release.
• Keep learning: bring ideas from AI, robotics, and systems engineering into practical products.

Programming languages

Backend, data and AI

Platforms and development tools

• Designing and improving dependable software products and developer tools.
• Exploring systems programming, concurrency, and performance in Rust and C++.
• Building practical AI applications with useful evaluation and retrieval workflows.
• Continuing research in robotics, reinforcement learning, and real-hardware systems.