AI • Software Engineering • Teaching
Engineering-first knowledge about AI, GenAI, RAG, evaluation, agentic systems, cloud architecture and software engineering—focused on architecture, failure modes, experiments, implementation and production readiness.
From learning to teaching.
Retrieval Quality & AI Evaluation • Article 02
A correct answer does not prove that retrieval worked. This article uses a reproducible experiment to separate retrieval correctness, evidence support, answer correctness, and the overall RAG result.
Read Article 0201
Chunking, hybrid retrieval, reranking, metadata, provenance and measurable retrieval quality.
02
Golden datasets, failure analysis, regression testing, acceptance criteria and evaluation-driven releases.
03
Agent boundaries, tool contracts, orchestration, deterministic services and Model Context Protocol.
04
Serverless AI systems, security, identity, observability, deployment and production architecture.
05
Testing, CI/CD, contracts, data models and the engineering disciplines AI systems still need.
06
Practical notes for engineers learning, experimenting, building and teaching modern AI systems.
“Terkeka First.
Everywhere Else Second.”
Terkeka is the canonical home for complete, maintained knowledge artifacts. Professional networks and technical communities help the knowledge travel—the source stays on Terkeka.
Terkeka