Odinzen is an independent research practice that builds open‑source scientific software and AI‑driven agentic systems for computational materials science, leveraging CALPHAD phase equilibria and thermodynamic modeling. Their platform integrates structured domain knowledge with real solvers, enabling rigorous, traceable workflows, and they also provide workshops and talks to help engineers adopt these methods in production environments.
Funding
Funding not disclosed
Founders
Product
Problem
Materials researchers often rely on fragmented tools and manual workflows to perform thermodynamic modeling and phase equilibria calculations, leading to reproducibility challenges and inefficient use of expert time.
Solution
Odinzen offers an independent research practice that provides computational materials science services built on open‑source scientific software and agentic AI. By integrating CALPHAD phase equilibria solvers with LLM‑driven agents, the platform automates workflow steps while continuously verifying results against domain knowledge. Structured knowledge graphs capture materials data, enabling machine reasoning and traceable AI‑driven analyses. Odinzen also delivers data pipelines, workshops, and consulting to help organizations adopt rigorous, reproducible AI‑enhanced materials research practices.
Target Audience
Primary customers are industrial R&D labs, academic research groups, and materials engineering teams seeking automated, reproducible thermodynamic modeling and AI‑enhanced workflow integration.
Features
- Open‑source CALPHAD assessments and thermodynamic property modeling tools integrated into a unified workflow
- Agentic AI system where LLM agents invoke real solvers and validate outputs against scientific rules and data
- Knowledge graph infrastructure for structured representation of materials data suitable for automated reasoning
- End‑to‑end data pipelines that ensure traceability and reproducibility of computational results
- Hands‑on workshops and consulting services to train teams in adopting AI‑augmented materials research methods