MiroMind Finalizes Scientific Leadership For Heavy Duty Solver Across Reasoning, Runtime, and Verifiable AI
REDWOOD CITY, Calif., March 12, 2026 - MiroMind has appointed three distinguished scientists to lead its core research pillars: reasoning models and training, runtime and agent systems, and verifiable AI. The goal is clear: build Discoverable Intelligence - AI that explores, proposes, and proves ideas, not just rephrases what already exists.
Founded by serial entrepreneur and philanthropist Tianqiao Chen, MiroMind is focused on rigorous reasoning and machine-checkable outputs for high-stakes domains. This is not a pursuit of a more eloquent language model. It's an engineering agenda for reliable solutions that stand up to scrutiny.
Dr. Simon Shaolei Du - Lead Scientist for Reasoning Models & Training
Dr. Du is an Associate Professor at the Paul G. Allen School of Computer Science & Engineering at the University of Washington. His work spans machine learning theory, deep learning optimization, and large-scale reasoning model training.
- Member of Technical Staff at xAI, contributing to large model R&D
- Visiting Professor at Facebook AI Research (FAIR)
- Postdoctoral Researcher at the Institute for Advanced Study, Princeton
At MiroMind, he will lead the end-to-end buildout of reasoning model architecture and training systems. Expect systematic gains in long-form reasoning, data curriculum, and training strategies that push reliability, not just benchmarks.
Prof. Bo An - Lead Scientist for Runtime & Agent Systems
Prof. An is a tenured Professor at Nanyang Technological University with a research track record in multi-agent systems, reinforcement learning, game theory, and AI decision-making. His work appears in leading venues including NeurIPS, ICML, and AAAI, and he has led multiple large-scale research initiatives while mentoring doctoral researchers.
He also brings experience transferring research into production across industry and government collaborations on decision systems, autonomous coordination, and large-scale optimization.
At MiroMind, he will oversee the runtime and agent execution layer, integrating reasoning models with the verification core. The objective: a horizontally scalable, auditable system with dependable multi-agent coordination and strong system-level reliability. For a broader view on tooling in this area, see AI Agents & Automation.
Dr. Kaiyu Yang - Lead Scientist, Verifiable AI Lab
Dr. Yang joins from Meta Fundamental AI Research (FAIR), following postdoctoral work at Caltech. He is known for contributions to verifiable reasoning and formal proof systems - a direct match with MiroMind's emphasis on machine-checkable correctness.
He will establish and run the Verifiable AI Lab with two core thrusts: verifiable reasoning and verifiable generation. Applied to code synthesis, mathematical proof, and complex reasoning tasks, the lab will drive a tight loop from research validation to engineered systems and product deployment, with guarantees that can be mechanically checked.
Why this matters for scientists and research teams
- Reproducibility you can audit: traceable chains of thought, explicit assumptions, and formal verification where possible.
- System-level reliability: agent coordination that is measured, testable, and instrumented for real workloads.
- From theorem to tool: research outcomes moved into production workflows with clear interfaces and guarantees.
- Open collaboration: follow progress and artifacts via GitHub and HuggingFace.
If your work depends on validated reasoning for science, engineering, or compliance, track this space closely. A practical overview of adjacent methods and tools is here: AI for Science & Research.
About MiroMind
MiroMind is an AI research and technology company in Redwood City, CA. The team is building a General Purpose Solver - a reasoning-first AI system engineered to be provably correct, not just plausible.
By advancing verifiable, long-chain System 2 reasoning, MiroMind targets high-stakes use cases across software engineering, financial services, healthcare and pharma, legal and compliance, and scientific research. The company was founded by Tianqiao Chen and is backed by a team that is 80%+ PhD researchers with a global scientific leadership bench.
Get involved
- Website: miromind.ai
- GitHub: github.com/MiroMindAI
- HuggingFace: huggingface.co/MiroMind-ai
- iOS App: App Store
- Android App: Google Play
- Media: press@miromind.ai
- Careers: talent@miromind.ai
Your membership also unlocks:
