cd ..

daita@system:~$ cat ./applied-ai.md

# Applied AI

Agentic systems, LLM pipelines, evals, retrieval.

## Focus

Pushing language-model systems past demo state into reliable, evaluated, observable production behavior.

## Research themes

  • Agent harnesses and tool use
  • Retrieval-augmented generation over messy enterprise data
  • Evaluation harnesses and failure-mode catalogs
  • Document extraction with source grounding
  • Cost-aware LLM routing and caching

## Public artefacts

  • Open-source eval scaffolds
  • RAG benchmarks across LanceDB, Qdrant, pg_vector
  • 30+ technical posts on agent and eval design

Have a research problem or a hard system to build? Talk to us.

Start a conversation →