Chief Technology Officer / 首席技術官

Deep Principle Limited; Hong Kong; Vor Ort; HKD 3.000.000 - 5.000.000; Vollzeit

Zusammenfassung

Deep Principle Limited is seeking a Chief Technology Officer in Hong Kong to lead Mira Global's AI models, agent systems, and computing infrastructure. You will own technical architecture, prototype development, and engineering delivery, shaping research plans and ensuring reproducible results at scale.

The role demands hands-on leadership, direct involvement in algorithms and system designs, and strong progress tracking across tool integration, workflow orchestration and scalable computing.

Qualifikationen

• Master's degree in computer science, AI, computational science, engineering, physics, chemistry or a relevant STEM discipline.
• 5 years of working experience in relevant field.
• Ability to lead integration across AI agent architectures, tool use and workflow orchestration.

Aufgaben

• Own technical architecture and experimental development. Translate scientific requirements into system designs and technical research plans.
• Prototype and compare approaches to resolve uncertainties in model integration and computational scalability.
• Develop scientific agent execution systems with tool calling, task orchestration and monitoring.
• Improve reliability and traceability across multi-model workflows and heterogeneous computing environments.
• Lead model training/inference capabilities, including pipelines, retrieval, deployment and cost trade-offs.

Kenntnisse

Python

Leadership

System architecture

Ausbildung

Master's degree in CS/AI/engineering or related STEM

Tools

PyTorch

TensorFlow

Jobbeschreibung

We are seeking a Chief Technology Officer to lead the research and experimental development of the models, agent systems and computing infrastructure underpinning Mira Global. The principal focus of this position is to resolve technical challenges in implementing scientific AI: how models use computational tools, how research workflows execute reliably, and how data and computing resources support reproducible results at scale. The CTO will translate scientific objectives into technical research plans and own technical architecture, prototype development and engineering delivery. This is a hands‑on technical leadership role requiring direct involvement in critical algorithms, system designs and code.

Key responsibilities

Own technical architecture and experimental development. Translate scientific requirements into system designs and technical research plans. Prototype and compare alternative approaches to resolve uncertainties in model integration, workflow execution and computational scalability.

Develop scientific agent execution systems. Research and implement mechanisms for tool calling, task orchestration, state management, execution monitoring and recovery from failed tasks. Improve reliability and traceability across workflows involving multiple models and scientific tools.

Advance model training and inference capabilities. Develop and evaluate methods for adapting models to scientific tasks, including training pipelines, retrieval and tool integration, inference optimisation and model serving. Measure trade‑offs between task quality, latency, throughput and computing cost.

Integrate scientific computing tools. Develop reusable interfaces connecting AI systems with first‑principles calculation packages, molecular simulation software and analysis tools. Address technical challenges in input validation, job scheduling, result parsing and execution across heterogeneous computing environments.

Build reproducible scientific data systems. Design data models and pipelines linking scientific inputs, computational settings, code and model versions, execution records and results. Develop mechanisms for provenance tracking, dataset versioning and reproducible workflow execution.

Establish system‑level evaluation and technical validation. Build benchmark environments and controlled experiments to assess agent task completion, tool‑use correctness, workflow reproducibility and system performance. Investigate failures and improve the underlying algorithms and architecture.

Lead research‑to‑product engineering. Integrate validated technical advances into Mira Global and customer research workflows. Set engineering standards for APIs, code review, testing, security and access control, and resolve architectural constraints affecting deployment and use.

Develop interfaces for future laboratory integration. Work with experimental teams on instrument connectivity, experimental data exchange and feedback interfaces needed for the planned computation–experiment loop.

Build and lead the Hong Kong technical team. Recruit and mentor AI, scientific computing, data and platform engineers. Remain directly involved in critical prototypes, implementation reviews and complex technical problems, and document original technical findings for intellectual property protection.

About you

Master's degree in computer science, artificial intelligence, computational science, engineering, physics, chemistry or a relevant STEM discipline with 5 years of working experience; Working experience is not compulsory for Doctoral degree holder.

Demonstrated experience leading the development of technically complex AI systems, scientific software or computing platforms, with clear evidence of personal contributions to architecture and implementation.

Strong programming skills in Python and practical experience with machine learning frameworks such as PyTorch or TensorFlow.

In-depth expertise in at least one of the following areas, together with the ability to lead integration across related systems: AI agent architectures, tool use and workflow orchestration; Model training, inference optimisation and deployment; High‑performance computing and automated scientific workflows.

Strong understanding of backend systems, APIs, databases, distributed computing and cloud or GPU infrastructure.

Ability to design controlled technical experiments, establish meaningful performance baselines and diagnose failures across models, data, software and computing infrastructure.

Experience building reproducible software and data workflows, with sound practices in version control, testing, documentation and technical review.

Ability to lead engineering teams while remaining actively involved in coding, architecture decisions and difficult technical problem‑solving.

Strong written and verbal communication skills in both English and Chinese, with the ability to translate scientific requirements into implementable and testable technical specifications.

About us

Deep Principle combines generative artificial intelligence and first‑principles calculations to accelerate chemical and materials innovation. Its Mira AI Scientist platform supports scientific research workflows spanning research planning, computational execution, analysis and iterative optimisation. Deep Principle Limited is building its Hong Kong R&D team to advance AI algorithms, scientific agents and model/platform engineering for Mira Global. Its development roadmap includes domain‑specific models, integration with international customers' research workflows, and subsequent development of wet‑laboratory capabilities connecting AI, computational research and experimental validation.