Overview
Skills
Job Details
Job Summary:
Client is seeking an Agentic (Autonomous) Systems Engineer to help bridge the gap between AI-generated code and production-ready autonomous systems. You will play a critical role in reviewing, refining, and integrating agent-generated outputs into scalable, safe systems. This position is ideal for engineers eager to grow in the emerging field of AI autonomy while developing strong fundamentals in backend development, tooling, and cloud infrastructure.
Engineer contributes directly to agent feedback loops, improves AI-driven workflows, and ensures features safely reach production. You will work alongside senior engineers to enhance agent capabilities, debug complex interactions, and strengthen the reliability of AI-driven processes.
**This is an on-site (hybrid 3:2) role in the Durham, NC office.**
Key Responsibilities:
Backend Development:
- Write and maintain production-quality backend services in Node.js, Java, and TypeScript using modular, testable architectures.
- Design APIs and services to integrate LLM-generated code into real-time pipelines and agentic workflows.
AI Code Validation & Integration:
- Review and sanitize LLM-generated code (e.g., tools, scripts, plugins), refactor for reliability, and integrate into production using secure coding practices.
- Extend agent execution environments with sandboxing, dependency management, and resource isolation (e.g., via Docker).
Agentic Workflow Engineering:
- Build and optimize pipelines for code generation, validation, and deployment using frameworks such as LangChain, Hugging Face Transformers, or internal orchestration systems.
- Implement automated monitoring hooks, runtime validation steps, and failure recovery mechanisms in multi-step AI workflows.
System Observability & Debugging:
- Instrument systems with Grafana, Prometheus, ELK stack, or equivalent observability tools for monitoring agent behavior, resource usage, and execution anomalies.
- Debug and diagnose complex runtime issues involving dynamic code execution, agent orchestration, or API/service failures.
CI/CD & Deployment:
- Contribute to containerized CI/CD pipelines using Docker, GitHub Actions, and cloud-native tools.
- Support staging and production deployment processes on AWS, Azure, or Google Cloud Platform, ensuring high reliability and compliance with security standards.
Security & Governance:
- Enforce secure handling of secrets, API keys, and credentials across agent executions and pipelines.
- Apply threat modeling to autonomous systems that handle sensitive data or customer-specific configurations.
Example Deliverables & Outcomes
- Build a Git-integrated module that enables agents to create PRs, manage issues, or sync code to repositories.
- Extend sandbox environments with third-party libraries or language runtimes and monitor for security violations.
- Design and execute validation pipelines that test LLM-generated scripts before deployment.
- Optimize agent workflows to reduce latency, improve success rates, and detect generation errors early.
- Create diagnostic dashboards that monitor agent executions, surface errors, and track success metrics over time.
Required Skills & Experience
- Must have 2-4 years of experience in software engineering with a focus on Agentic AI Systems, and backend or systems development.
- Proficient in Node.js, Java, and TypeScript; strong software design and debugging skills.
- Experience building and maintaining RESTful APIs, backend microservices, or developer tooling.
- Solid understanding of algorithms, data structures, and scalable systems design.
- Exposure to CI/CD workflows, Docker-based build pipelines, and code release practices.
- Hands-on experience with at least one major cloud platform (AWS, Azure, or Google Cloud Platform).
- Familiar with Git, branching strategies, and collaborative development workflows.
Qualifications (Required)
- Experience with LLM tools and frameworks (e.g., LangChain, OpenAI APIs, Hugging Face).
- Basic familiarity with Kubernetes, Helm charts, or service orchestration.
- Knowledge of AI agent safety practices, dynamic execution risks, and code validation strategies.
- Working knowledge of observability tools: Grafana, Prometheus, Kibana, or similar.
- Understanding of data privacy, secret management (e.g., Vault), and secure API access patterns.