AI Security Engineer

Agentic AI Developer and Integration Engineer

About the job

CyForce is a cybersecurity company focused on delivering advanced security solutions, offensive capabilities, and practical innovation to its customers.

We are looking for an *Agentic AI Developer and Integration Engineer* to design, build, and integrate intelligent AI agents into a new customer-facing cybersecurity product.

This is not a traditional AI application development role.

We are looking for a builder who understands how to move beyond simple chat-based interfaces and create a truly agentic product - a system that can plan, reason, use tools, coordinate tasks, analyze technical evidence, validate results, and operate within controlled security workflows.

You will be responsible for developing the AI-agent layer of the product and integrating it with security tools, backend systems, data sources, and operational workflows.

This is an opportunity to help define how Agentic AI can be applied to real-world cybersecurity problems and to build a product that combines intelligent automation with practical security capabilities.

What you’ll do

* Design and develop the agentic architecture of a customer-facing cybersecurity product

* Build autonomous and semi-autonomous AI agents capable of planning, executing, validating, and documenting technical tasks

* Develop single-agent and multi-agent workflows for cybersecurity use cases

* Integrate LLMs with security tools, APIs, scanners, databases, cloud services, sandboxes, and internal systems

* Develop tool-calling, memory, retrieval, context management, planning, orchestration, and structured-output capabilities

* Build agent workflows for research, reconnaissance, analysis, security validation, test planning, and technical reporting

* Define agent permissions, execution boundaries, approval flows, and rules of engagement

* Develop secure integrations between AI agents and offensive security capabilities

* Implement human-in-the-loop controls for high-impact or sensitive actions

* Build validation mechanisms to identify hallucinations, tool failures, false positives, incomplete execution, and inconsistent outputs

* Develop agent observability, tracing, logging, evaluation, and performance-monitoring capabilities

* Improve agent reliability, accuracy, scalability, and operational consistency

* Integrate multiple AI models and select the appropriate model based on task, context, cost, and risk

* Build connectors and adapters for external tools, APIs, data sources, and customer environments

* Collaborate with Offensive Security specialists, software developers, product stakeholders, researchers, and R&D teams

* Translate customer requirements and cybersecurity workflows into scalable agentic product capabilities

* Research emerging developments in AI agents, LLMs, orchestration frameworks, AI security, and cybersecurity automation

What you’ll bring

* Strong programming experience in Python

* Hands-on experience building AI agents, LLM applications, multi-agent systems, or agentic workflows

* Experience integrating LLMs with external tools, APIs, databases, or production systems

* Strong understanding of tool and function calling, structured outputs, RAG, embeddings, memory, and workflow orchestration

* Experience with agent planning, task decomposition, state management, context management, and execution control

* Experience building backend services, integrations, or production-grade AI applications

* Familiarity with cybersecurity concepts and an interest in Offensive Security

* Understanding of Web Applications, APIs, networking, authentication, infrastructure, or cloud environments

* Experience working with Linux, containers, APIs, queues, databases, and modern development environments

* Strong analytical skills and the ability to diagnose complex model, agent, integration, and infrastructure failures

* Understanding of the risks involved in granting AI systems access to tools, sensitive data, and customer environments

* Ability to design systems that balance autonomy, security, reliability, and operational control

* Strong ownership, independent research capabilities, and product-oriented thinking

Bonus points

* Hands-on experience in Offensive Security, Penetration Testing, AppSec, or Security Research

* Experience building AI agents for cybersecurity, DevSecOps, IT operations, or technical automation

* Familiarity with LangGraph, LangChain, AutoGen, CrewAI, Semantic Kernel, MCP, or similar technologies

* Experience integrating multiple models, local models, open-source models, or model-routing systems

* Experience with Azure OpenAI, AWS Bedrock, Google Vertex AI, or similar AI platforms

* Knowledge of Prompt Injection, Indirect Prompt Injection, Data Leakage, Excessive Agency, insecure tool usage, and AI supply-chain risks

* Familiarity with the OWASP Top 10 for LLM Applications and other AI security frameworks

* Experience with agent evaluation, benchmarking, red teaming, observability, tracing, and prompt-version management

* Experience with Docker, Kubernetes, CI/CD, DevSecOps, distributed systems, or sandboxed execution environments

* Experience building connectors, plugins, tool adapters, or orchestration layers

* Understanding of Web Security, API Security, Active Directory, Cloud Security, or attack-path analysis

* Experience developing customer-facing SaaS or cybersecurity products

* Relevant certifications in AI, Software Development, Cloud, or Offensive Security

If you are excited about building a truly agentic cybersecurity product, connecting AI agents to real tools and operational workflows, and helping CyForce create a new AI-driven security capability for its customers - we would like to hear from you.