桌面端Agent Sandbox研发工程师/专家 - 飞书
Core
Develop secure sandbox environments for AI Agents across Windows, Linux, and macOS, providing process control, file isolation, and network access management.
Role type
Senior IC systems engineer (sandbox security)
Builds
Secure execution environments for AI agents
Domain
Cybersecurity + Systems Programming
Deliverable
production ML models
Required skills
C/C++, binary security, process/thread management, memory management, reverse engineering, debugging (Windbg/GDB/LLDB), system call interception, eBPF, kernel/driver development
Preferred skills
Windows NT architecture, injection/anti-injection techniques, Linux namespaces/cgroups, macOS XNU/dyld, container runtime security, virtualization (KVM/QEMU)
Technologies
C/C++, GDB, strace, perf, valgrind, IDA Pro, Ghidra, eBPF, KVM, QEMU, Win32, COM/DCOM, LSP/NSP, Seccomp, Landlock, TCC, SIP, XPC
Responsibilities
Design and implement cross-platform sandbox architecture; build security policies for process, file, and network isolation; optimize stability and performance; analyze crashes and memory leaks; develop reverse engineering capabilities for system analysis.
Seniority
Senior, hands-on IC