Systima
  • Agentic AI PlatformsProduction-grade multi-agent systemsFractional Head of AISenior AI leadership on demandAI Governance & ComplianceEU AI Act and regulatory frameworksAI Strategy & RoadmappingFrom vision to execution planLLM Cost OptimisationCut LLM spendAI Due DiligenceTechnical AI assessment for investors
  • AI for LegalTechAI systems for legal workflowsAI for SaaS & B2BEmbed AI into your product
  • About
  • Blog
Agentic AI PlatformsFractional Head of AIAI Governance & ComplianceAI Strategy & RoadmappingLLM Cost OptimisationAI Due Diligence
AI for LegalTechAI for SaaS & B2B
AboutBlog

LLM Engineering

Practical techniques for working with large language models.

AllAI GovernanceAI LeadershipAgentic AIApplied ResearchLLM EngineeringRegulated IndustriesStrategy
LLM Engineering

The Delimiter Hypothesis: Does Prompt Format Actually Matter?

600 model calls across four frontier LLMs testing XML, Markdown, and JSON delimiters. Format usually does not matter. MiniMax M2.5 is the exception.

5 March 202617 min read
AI GovernanceLLM Engineering

Open-Source Article 12 Logging Infrastructure For The EU AI Act

An open-source audit logging library for the EU AI Act. What it does, what it deliberately does not, and why that gap matters more than the code.

1 March 20268 min read
AI GovernanceLLM Engineering

What to Log, How Long to Keep It, and How to Reconstruct: Article 12 for Engineers

Article 12 is the most technically demanding obligation in the EU AI Act. Given a decision ID from six months ago, can you reconstruct the full chain?

27 February 202615 min read
AI GovernanceLLM Engineering

Your Model Is in Production. Now What? Post-Market Monitoring Under Article 72

Most engineering teams treat deployment as the finish line. Article 72 makes it a starting point. Post-market monitoring is a legal obligation.

27 February 202610 min read
Systima© 2026 Systima. All rights reserved.United Kingdom · European Union
AboutServices