Beyond RAG: A Relational Context Engine That Cuts Token Burn — Unblocked
Unblocked's relational context engine delivers task-specific, interconnected context to LLMs, cutting token waste versus naive RAG.
“Here I spent, let's say, $1.29, and it took about a minute and thirty”
Unblocked demos a relational context engine that aggregates code, Notion docs, pull requests, and Slack into a structured, permission-aware graph for LLMs, replacing brute-force codebase crawling. The demo shows Claude completing an optimization task with pre-assembled cross-linked context rather than iterating blindly through a repo, with a cost comparison hinting at significant token reduction. Relevant to teams building agentic developer tooling, though the talk is cut off before hard numbers are revealed.