AI Security Engineer, Vulcan

We are seeking a highly skilled and experienced Prompt and AI Security Engineer to address the most pressing customer concerns regarding AI security, prompt engineering, and model integrity. The ideal candidate will have hands-on development experience, the ability to integrate and customize security solutions for target endpoints (applications or models), and expertise in defining and refining prompts generated by AI systems. This role involves working closely with clients to mitigate security risks, ensure AI safety, and contribute to the evolution of robust AI security solutions.

我們正在尋找一位具備豐富經驗的 AI 資安工程師，
專注於解決客戶在AI安全、提示工程（Prompt Engineering）與模型完整性 方面的關鍵問題。

該職位應具備實際開發經驗，能夠針對目標端點（應用程式或模型）整合並客製化資安解決方案，並精通AI架構與提示詞（Prompt）優化與調整。
此職位將與客戶密切合作，協助降低AI安全風險，確保AI運作的安全性，
並推動AI資安解決方案的發展與演進。

Vulcan: https://vulcanlab.ai/

Cymetrics: https://cymetrics.io/zh-tw/products/ai-redteam

-

How to apply

It will help us process your applications faster https://grnh.se/f279e1654us

*Please apply with English CV, thank you.

-

Responsibilities

Prompt Engineering and Security Optimization:

• Design, refine, and optimize AI-generated prompts to improve accuracy, safety, and efficiency.
  設計、調整並優化 AI 生成的提示詞，提升準確性、安全性與效率。
• Develop techniques to prevent prompt-based security vulnerabilities such as injection attacks and model manipulation.
  開發技術來防範基於 Prompt 的安全漏洞，例如注入攻擊與模型操控。
• Ensure system-generated prompts align with security and ethical AI guidelines.
  確保系統生成的 Prompt 符合資安標準與 AI 倫理規範。

Development and Integration:

• Customize and integrate AI security solutions with target endpoints, including applications and machine learning models.
  客製化並整合 AI 資安解決方案，適用於應用程式與機器學習模型。
• Implement security protocols and best practices for AI model deployment and interaction.
  實作 AI 模型的安全性協議與最佳部署實踐。
• Develop tools and automation pipelines to streamline prompt security testing and optimization.
  開發工具與自動化流程，強化 Prompt 安全測試與優化效率。

Quantitative Analysis and Risk Mitigation:

• Utilize text quantification techniques to define necessary security indicators and performance benchmarks.
  運用文本量化技術，定義必要的安全指標與效能標準。
• Conduct security assessments to identify vulnerabilities and implement mitigation strategies.
  進行資安評估，識別漏洞並制定對應的防禦策略。
• Analyze real-world AI security threats and develop response frameworks.
  分析 AI 相關的真實世界安全威脅，建立應對框架。

Client Engagement and Solutions Delivery:

• Collaborate with clients to understand their AI security concerns and provide tailored mitigation solutions.
  與客戶密切合作，深入了解 AI 資安需求並提供客製化風險緩解方案。
• Act as a technical advisor, offering best practices on secure AI implementation and deployment.
  擔任技術顧問，提供 AI 安全性實施與部署的最佳實踐建議。
• Conduct workshops, training, and documentation to support client adoption of AI security solutions.
  舉辦工作坊、培訓與撰寫技術文件，協助客戶導入 AI 安全解決方案。

Research and Thought Leadership:

• Stay updated on AI security trends, risks, and emerging threats, particularly in the context of LLMs and NLP models.
  持續關注 AI 安全趨勢、風險與新興威脅，特別是在 LLM（大型語言模型）與 NLP（自然語言處理） 領域。
• Explore novel verification techniques to ensure AI models remain secure, robust, and compliant.
  探索新的 AI 驗證技術，確保模型安全、穩定並符合法規要求。
• Contribute to internal research and development efforts to improve AI security methodologies.
  參與內部 AI 資安研究與開發，推動安全性技術進步。
• Optionally, write technical papers and articles to share findings and innovations.
  （選填）撰寫技術論文與文章，分享研究成果與創新技術。

-

Education Background:
Bachelor's, Master's, or Ph.D. in Computer Science, Artificial Intelligence, Machine Learning, Cybersecurity, or a related field. Equivalent industry experience will also be considered.
計算機科學、人工智慧、機器學習、資安或相關領域的學士、碩士或博士學位。具備同等業界經驗者亦可。

Technical Skills:

• Strong programming expertise in Python and experience with ML frameworks such as PyTorch, TensorFlow, or similar.
  精通 Python，並具備 PyTorch、TensorFlow 或類似機器學習框架的開發經驗。
• Hands-on experience with prompt engineering, NLP, NLU, and LLM fine-tuning.
  具備 Prompt Engineering、自然語言處理（NLP）、自然語言理解（NLU） 及大型語言模型（LLM）微調的實作經驗。
• Experience in securing AI models against adversarial threats, prompt injection attacks, and data poisoning.
  有 AI 資安防護經驗，包括對抗式攻擊、Prompt 注入攻擊、資料投毒等。
• Strong knowledge of cybersecurity principles, risk assessment, and AI-specific security methodologies.
  熟悉資安原則、風險評估 及 AI 相關的安全方法論。
• Familiarity with MLOps tools such as Kubeflow and MLflow.
  了解 MLOps 工具（如 Kubeflow、MLflow）並具備實作經驗。
• Understanding of secure software development practices and regulatory frameworks (e.g., GDPR, CCPA, AI Act).
  熟悉安全軟體開發實踐，以及相關法規（如 GDPR、CCPA、AI Act）。

Experience:

• Proven track record of developing, deploying, and optimizing ML and LLM-based systems.
  具備開發、部署、優化 機器學習及 LLM 系統的成功經驗。
• Hands-on experience integrating AI security solutions into production environments.
  有 AI 安全解決方案進入生產環境的整合經驗。
• Experience working with interdisciplinary teams (e.g., legal, compliance, or policy) to address AI safety concerns.
  具備與 跨領域團隊（如法務、合規、政策） 合作，解決 AI 安全風險的經驗。
• Client-facing experience, including presenting security solutions and conducting risk assessments.
  具備客戶對接經驗，包括安全解決方案展示、風險評估 等。

Soft Skills:

• Strong problem-solving and critical-thinking abilities.
  優秀的問題解決與批判性思維能力。
• Excellent communication skills, with the ability to convey technical concepts to diverse audiences.
  出色的溝通能力，能夠向不同背景的受眾清楚表達技術概念。
• Ability to write and speak in English fluently.
  流利的英文書寫與口語表達能力。
• Passion for developing robust, secure, and ethical AI systems.
  熱衷於開發安全、穩健且符合倫理的 AI 系統。
  

Nice-to-Have:

• Experience with AI model interpretability and explainability techniques.
  熟悉 AI 模型可解釋性與可解釋 AI 技術。
• Knowledge of federated learning, differential privacy, and secure AI training methodologies.
  了解聯邦學習、差分隱私、安全 AI 訓練方法。
• Background in AI compliance and auditing.
  具備 AI 合規性與審計相關經驗。

• Phone interview: 1 hour meet with HR
• Onsite Interview: 1-2 hours
  • 1-1.5 hours meet with our team and hiring managers
  • 0.5 hour meet with HR