Thiết kế và phát triển tính năng Agentic AI hỗ trợ phân tích và điều tra sự cố an ninh mạng.
Yêu cầu
Từ 3 năm (Mid) hoặc từ 5 năm (Senior) kinh nghiệm phát triển phần mềm/AI
Ít nhất 1 năm kinh nghiệm xây dựng và vận hành ứng dụng LLM trên production
Thành thạo Python, thiết kế API, làm việc với database và triển khai dịch vụ
Kinh nghiệm thực hành với RAG, vector search, prompt engineering, agent workflows
Khả năng đọc hiểu tài liệu kỹ thuật và giao tiếp tốt bằng tiếng Anh
Trách nhiệm
Thiết kế hệ thống LLM: retrieval, quản lý context, tool calling, structured output
Phát triển AI agent hỗ trợ truy vấn log bằng ngôn ngữ tự nhiên và tóm tắt điều tra
Xây dựng bộ dữ liệu đánh giá cùng Data Scientist và regression check
Cải thiện chất lượng, độ trễ, chi phí vận hành và triển khai monitoring cho production
Có thì tốt
Hiểu biết về vận hành SOC, điều tra sự cố, threat hunting hoặc phân tích log bảo mật
Kinh nghiệm xử lý rủi ro AI như prompt injection, rò rỉ dữ liệu
Kinh nghiệm xây dựng workflow chạy dài hạn có lưu trạng thái và phục hồi lỗi
Tóm tắt bằng AI từ JD gốc, đọc JD gốc trước khi ứng tuyển.
Job description
We are looking for a Senior/Middle AI Engineer to build AI capabilities for our products, helping security teams explore data, analyze alerts, and investigate incidents more effectively.
You will design, implement, and deliver AI features while providing technical direction within your area of ownership. This role suits someone with strong software engineering fundamentals, practical experience building LLM applications, and an interest in solving real cybersecurity problems.
Key responsibilities
Design and build LLM-powered systems, including retrieval workflows, context management, tool calling, and structured outputs.
Develop AI agents that support natural-language log queries, evidence collection, and investigation summaries grounded in source data.
Define appropriate automation boundaries and introduce human approval for sensitive actions.
Integrate AI capabilities with product APIs, backend services, and data sources.
Partner with Data Scientists to build evaluation datasets from real scenarios, analyze failures, and establish regression checks for model, prompt, and workflow changes.
Improve response quality, latency, and operating costs; implement monitoring and diagnostics for production systems.
Implement access controls for agent tools and data, with reliable handling of missing evidence, invalid queries, and tool failures.
Review technical designs and code, mentor AI Engineers, and develop reusable components for the team.
Collaborate with Product, Backend, and security specialists to translate operational needs into features with clear acceptance criteria.
Qualifications
Education: Bachelor’s degree/ Master’s degree or in Computer Science, Artificial Intelligence, Machine Learning, Electrical Engineering, or a related field.
3+ years of experience in software development or AI engineering for Mid-level, or 5+ years of experience for Senior-level, including at least 1 year building and operating LLM applications in production.
Strong Python skills and experience designing APIs, working with databases, and deploying services.
Hands-on experience with RAG, vector search, prompt engineering, and agent workflows.
Ability to design solutions from requirements through deployment and explain trade-offs involving reliability, performance, and cost.
A systematic approach to evaluating AI quality using data and experiments.
Demonstrated ownership across the feature lifecycle: requirements, implementation, testing, release, and continuous improvement.
Clear communication, proactive problem-solving, and the ability to mentor teammates.
Ability to read technical documentation and communicate effectively in English.
Nice to Have:
Familiarity with SOC operations, alert triage, incident investigation, or threat hunting.
Experience processing security logs, integrating SIEM/SOAR platforms, or building tools for security analysts.
Experience addressing prompt injection, data leakage, or risks associated with agent tool use.
Experience building long-running workflows with state persistence, failure recovery, and execution tracing.
Open-source contributions or projects that demonstrate practical AI engineering skills.
