FuSaExperts is training and fine-tuning LLMs specifically for safety engineering - because general-purpose LLMs are not reliable enough for standards-driven work. In functional safety, it’s not about sounding correct; you need consistent application of safety principles, best practices, and repeatable, review-ready outputs across artifacts like Safety Plan, HARA/risk analysis, FSC/TSC, impact analysis, and change management.
Our approach uses fine-tuning methods that match the role and the work package:
- SFT (Supervised Fine-Tuning) teaches correct structure, terminology and expected outputs: templates, content patterns and lifecycle logic.
- DPO (Direct Preference Optimization) aligns the model with safety-quality preferences: clarity, traceability, conservative assumptions and fewer hallucinations.
- CFT (Critique Fine-Tuning) hardens performance through targeted critiques and revisions - especially for reviewer behavior and recurring failure modes.
We apply these methods differently depending on whether the model acts as a Coach, Developer or Reviewer, and depending on the exact safety task, such as safety planning, risk analysis, impact analysis or safety concept development.
The workflow has three phases: principles and best-practice training using SFT and DPO; production of structured, traceable and auditable safety work products; and a feedback loop using CFT on real outputs and review findings to continuously improve quality, consistency and reviewer readiness.
← More insights