AI Knowledge & Learning Systems | Search, RAG & LLM Evaluation | Technical Learning & Developer Education
I build systems that help people find, understand, practise, and apply complex technical knowledge.
My background combines 13+ years in technical support and application engineering with deep experience in Elasticsearch, search/retrieval, production diagnostics, observability, knowledge workflows, and technical problem solving.
I'm currently exploring how those capabilities can be applied to:
- AI-powered knowledge and learning systems
- RAG, hybrid retrieval, semantic search, and reranking
- LLM evaluation, reliability, and guardrails
- knowledge-to-learning pipelines
- adaptive technical learning
- conversational and voice-based learning
- technical simulations and diagnostic reasoning
- developer education and AI-assisted expert workflows
The common thread across my work is measurable quality: grounded answers, provenance, retrieval metrics, reproducible diagnostics, evaluation datasets, human review, and systems that help people build understanding rather than simply generate answers.
Cross-platform learning application for German practice across TestDaF, workplace communication, and negotiation.
It combines conversation and writing practice, personal learning materials, audio, vocabulary, saved progress, speech recognition, voice interaction, and AI-assisted feedback.
What I'm exploring
- conversational AI for learning
- voice and speech interfaces
- personalised practice
- learner progress
- local-first/private learning workflows
- how AI can support practice without replacing the learner's own thinking
Production-shaped RAG system over technical documentation using Elasticsearch hybrid retrieval, reranking, citations, feedback, evaluation, and observability.
What it demonstrates
- grounded generation
- measurable retrieval quality
- provenance and citations
- LLM observability
- human review
- insufficient-evidence handling
This forms part of the knowledge infrastructure I want to apply to adaptive learning systems.
Operational knowledge-quality system for identifying duplicate and overlapping technical articles, clustering related knowledge, and supporting editorial review.
What it demonstrates
- knowledge quality
- semantic similarity
- human-in-the-loop AI
- evaluation against human decisions
- safe agent workflows
- knowledge governance
Interactive technical quiz focused on engineering judgement across Elasticsearch, distributed systems, observability, and resilience.
Each problem includes explanations and evidence from official documentation.
This is an early experiment in turning technical knowledge into active practice rather than passive documentation.
→ elasticsearch-resilience-quiz
E-commerce product search relevance lab with Elasticsearch mappings, deterministic ingestion, BM25 strategy comparison, relevance metrics, latency benchmarks, and local search quality gates.
What it demonstrates
- measurable search relevance
- ranking comparison with Precision@5, MRR@10, and nDCG@10
- ingestion quality and its effect on search
- reproducible latency benchmarks and quality gates
- elastic-search-policy-control-plane — deterministic search policies and explainable query execution.
- elastic-repo-inventory — provenance-aware technical retrieval and version-aware search.
- elastic-ai-search-decision-lab — documentation findability evaluated with practitioner questions and relevance metrics.
I'm particularly interested in moving from:
documentation → retrieval → answers
toward:
knowledge → diagnosis → practice → feedback → evaluation → mastery
Upcoming experiments include:
- an AI troubleshooting simulator
- adaptive technical tutoring with progressive hints
- knowledge-to-learning pipelines
- learner-state and misconception modelling
- evaluation of AI tutoring quality
- conversational simulation
- small immersive/WebXR learning experiments
AI & retrieval: RAG, hybrid search, semantic search, vector search, reranking, LLM evaluation, agents
Search: Elasticsearch, BM25, relevance evaluation, Precision@k, MRR, nDCG
Engineering: Python, FastAPI, TypeScript, Node.js, Docker
Observability: OpenTelemetry, structured logging, tracing, metrics
Knowledge: KCS, knowledge quality, deduplication, provenance, human review
I prefer systems that:
- make complex knowledge understandable
- expose evidence rather than hide it
- measure quality instead of assuming it
- keep humans in control where judgement matters
- turn recurring problems into reusable knowledge and tools