About the job

Software Engineer

Location: Germany | Remote | Full-time

A well-funded, early-stage cybersecurity company is looking for a Software Engineer — Agent Guardrails to own the runtime instrumentation layer that enables real-time security and policy enforcement for AI agents.

This is a highly technical opportunity focused on building the sensors that observe what AI agents are doing at runtime, generating events that allow the policy engine to make real-time block or allow decisions.

You’ll design and build low-latency runtime instrumentation across Node.js and Python, develop hooks for emerging AI protocols and SDKs such as MCP, and build tooling that automatically detects changes in fast-moving third-party dependencies.

Ideal candidates will have:

• Strong production experience with Node.js and Python

• Hands-on experience with dynamic module loading, monkey-patching, hooks, interceptors, and runtime instrumentation

• Strong understanding of asynchronous execution, including asyncio and the Node.js event loop

• Experience working with FFI and compiled languages such as Rust, C, or C++

• Understanding of memory management, thread safety, and dynamic linking

• Experience building resilient software on top of fast-evolving third-party SDKs and runtimes

• Strong debugging skills, including race conditions, memory leaks, and silent runtime failures

• Experience designing low-latency, performance-sensitive systems

• Strong ownership mentality and ability to work independently in an early-stage environment

Desirable Experience:

• AI agent infrastructure and LLM SDKs

• MCP (Model Context Protocol) or other emerging AI protocols

• Runtime security, application security, or cybersecurity

• Web frameworks and execution transports

• Observability, tracing, instrumentation, or developer tooling

• Experience working with rapidly evolving open-source ecosystems

Key Words:

Software Engineer / Systems Engineer / Runtime Engineer / Security Engineer / Python / Node.js / JavaScript / TypeScript / Runtime Instrumentation / Monkey Patching / Interceptors / Hooks / Asyncio / Node Event Loop / FFI / Rust / C / C++ / Dynamic Linking / Memory Management / Thread Safety / AI Agents / LLM / LLM SDK / MCP / Model Context Protocol / Runtime Security / Application Security / Cybersecurity / Observability / Distributed Systems / Performance Engineering

(By applying to this role you understand that we may collect your personal data and store and process it on our systems. For more information please see our Privacy Notice: Privacy Notice)