Scientific Consultant: AI for terminal device / AI Agent

Helsinki, Finland

Apply now Apply later

The Helsinki Systems Security Laboratory in Huawei (HSSL) drives renewal and mastery in the field of platform / device related security technologies for the mobile device. Our topical expertise is in hardware-assisted isolation and system protection (hypervisor, TEE, kernel hardening) as well as functions like device key management, attestation and integrity. 

The consultant will be part of a team developing machine learning (agent) solution for terminal devices. Below are the main responsibilities for this position, and the consultant will take part in one or more of these;

1) Optimizing the performance of the model usage by improving timing and memory use without losing too much accuracy, precision or recall of the model. 

2) Implementing novel use case such as file encryption mechanism, where file encryption/decryption keys are available only when device state has desired classification (determined by the agent). One possibility is to run unsupervised learning for clustering different device state categories as a device state labels. After clustering the AI agent can use model to classify current device state.

3) System security design for AI agent such as AppAgent (https://github.com/mnotgod96/AppAgent) to perform malicious actions. One possibility is to implement Guardian Agent (GA) that can analyze real time behavior of the AI agent. When GA detects that AI agent is tries high risk action it should stop action permanently or it could ask user consent to perform such action. The GA can inform OS to limit the capabilities of AI agent in order to stop malicious or high-risk action. 

We are looking for person who:

  • Has experience on optimizing AI models, such GPT models.
  • Has knowledge on some isolation technologies such as Arm Trustzone, Intel SGX/TDX, etc.
  • Has understanding about the security threats of AI models.
  • Has experience on analyzing and visualizing datasets.
  • Has some experience with C/C++ or Rust.
  • Has experience on Android development and preferably some experience on Android Native Development.
  • Has good English language skills.

Person we are looking should be able to help in the following tasks:

  • Provide clear guidelines for optimizing GPT models running in Secure Enclaves.
  • Prototype of some selected use case (such as device state classifier).
  • Prototyping “AgentGuard” solution and demonstrate how it can prevent malicious actions made by local AI Agent (such as multi-modal AppAgent)

About Huawei:

Founded in 1987, Huawei is a leading global provider of information and communications technology (ICT) infrastructure and smart devices. We have 207,000 employees and operate in over 170 countries and regions, serving more than three billion people around the world. We are committed to bringing digital to every person, home and organization for a fully connected, intelligent world. Scientific exploration and technological innovation drive civilization and society forward. Huawei understands the importance of research and innovation and how openness is critical for both. We are ready and willing to work with academia and industry to explore the frontiers of science and technology, push innovation forward, create value for industry and society as a whole, and build a better intelligent world. Learn More about Huawei.

Apply now Apply later
  • Share this job via
  • 𝕏
  • or

* Salary range is an estimate based on our AI, ML, Data Science Salary Index 💰

Job stats:  0  0  0

Tags: Classification Clustering GitHub GPT Machine Learning Prototyping Research Rust Security Unsupervised Learning

Region: Europe
Country: Finland

More jobs like this