Agentic Loops

Talk

When Models Become the Commodity

Abstract

ACP Engineering works extensively with LLMs to write production code across a small team. What makes that work is not the models, but the groundwork laid before them: a portfolio of Python projects sharing tooling, structure, tests, CI, and linters. That foundation powers ACP-M, its industrial IoT platform, and an agentic-loop prototype. This talk explains how fragmented operational data can become knowledge graphs, unified namespaces, and typed semantic layers so agents act on focused, reliable information.

Talk

Shipping Faster, Drowning Sooner

Abstract

AI ships features faster than ever, but it also ships bugs faster. Victor will show how LogicStar finds bugs across Sentry, Jira, and codebases, separates signal from noise, and helps teams fix issues that matter.

Talk

Practical Multi-Repo Agentic Engineering

Abstract

Practical techniques for reducing friction when doing agentic engineering across multiple repositories with VS Code and GitHub Copilot, including concurrent agent sessions, GitHub MCP workflows, and using Git history to improve agent instructions.

Talk

SLayer: the Dynamic, Agent-Native, Open Source Semantic Layer

Abstract

Most semantic layers were built for dashboards, not agents. SLayer is a lightweight, embeddable, open-source semantic layer that lets agents define complex queries using natural semantics, with MCP, CLI, REST, and Python APIs plus hybrid search over memories and entities.