//Areas

Lab areas

Six fronts of the same problem: putting AI agents to work without losing control. The first one is the lab's signature.

//01flagship

AI Ops Sec

Agent security and observability

red teamingMCPOWASP ASI2 articles
//02

Agents and orchestration

How to build an agent that survives production

MCP serversn8nmulti-agent1 articles
//03

Automation and ROI

Business processes handed to agents, with the numbers

processescostsmetrics0 articles
//04

Evaluation and quality

Knowing whether the agent got it right before the client does

evalsLLM-as-judgeregression0 articles
//05

Governance and regulation

What the law already asks of agents, read by an operator

AI ActNIS2ISO 420010 articles
//06

Models and infrastructure

Local and commercial models, and the infrastructure that serves them

local modelsMLXfine-tuning0 articles
//07

Tokenization

The token as the economic unit of AI: what it costs, what it saves, what it hides

AI FinOpscachetokenizers7 articles