§ Capabilities
Skills
Grouped by what they are for rather than listed alphabetically — the grouping is the point. Open as many as you want to compare.
Establishing that software does what it claims — by proof where the domain allows it, and by disciplined measurement everywhere else.
- Kani (bounded model checking)
- Verus (linear ghost types)
- Miri (undefined-behaviour detection)
- Property and invariant design
- Reference-equivalence testing
- Deterministic PRNG design
- Playwright end-to-end testing
- Jest / Vitest
- Mechanical quality gates
Fine-tuning small models for domains with scarce data, and building the agent systems and evaluation harnesses around them.
- LLM fine-tuning (Unsloth, Hugging Face Transformers)
- Qwen, Llama 3.2, Granite, DeepSeek
- OpenAI Agents SDK
- LangChain / LangSmith
- Model Context Protocol (MCP)
- Multi-agent architectures
- Harness-level evals and graders
- Hallucination detection
- Retrieval-augmented generation
- Vector stores
- Azure OpenAI
- Anthropic Claude, OpenAI GPT
Formal schemas that give a reasoner something to traverse and a model something to be held to.
- TypeDB / TypeQL
- Agentic and domain ontologies
- Ontology-driven code generation
- LinkML
- Semantic web (Jena, RDF)
- Datalog stores
Working proficiency across systems, application, and statistical languages.
- Rust
- TypeScript / JavaScript
- Python
- Kotlin
- SQL
- COBOL / JCL
- PHP
- Java
- R / SAS
- OCaml / ReasonML
Full-cycle application development, from design system through to production deployment.
- React
- Next.js
- TanStack Start
- Angular
- Ionic / Capacitor
- Three.js
- React Three Fiber
- Astryx
- Radix UI
- Tailwind CSS
- Drizzle ORM
- better-auth
- GraphQL / Apollo
- Express.js
Relational, document, graph, and vector stores.
- PostgreSQL
- TypeDB
- MySQL / MariaDB
- MongoDB
- pgvector
- Sequelize / Mongoose
Running services in production — provisioning, deployment, monitoring, and the pipelines between them.
- Docker
- Ansible
- nginx
- systemd
- VPS administration
- AWS
- Azure
- Vercel
- CI/CD pipelines
- Turborepo / pnpm
- Sentry
A master's degree, and the discipline underneath the numerical work.
- Hypothesis testing
- Bayesian statistics
- Resampling and bootstrap
- Regression and regularisation
- Clustering and decomposition
- Change-point detection
- Predictive modelling
- Time series analysis