Senior Agentic AI Developer

About Us And The Role

Cyberillium offers high-end security research and R&D services for customers who want to build secure products. We conduct in-depth research, architecture design, and low-level development across a wide range of technologies.

Founded by an elite group of Israeli security researchers, we deliver cutting-edge cyber solutions to enterprises of all sizes and the Israeli government. In addition to our services group, we are developing a unique cybersecurity product designed to bring a completely new approach to the cyber market.

As our Senior Agentic AI Developer, you will own the architecture and technical direction of our Agentic AI Malware Analysis and Threat Intelligence System: a mission-critical, on-premises platform designed for high-security, air-gapped environments. You’ll own the code, staying 100% hands-on, driving how the system is both architected and brought to life in production.

You will drive the integration of Large Language Models (LLMs), multi-agent orchestration frameworks (ReAct-style), and autonomous agents with high-fidelity malware analysis toolchains, enabling automated reverse engineering, behavioral analysis, and threat attribution at scale.

Requirements:

• Own the end-to-end architecture of a complex, multi-component system, including LLM pipelines, agent orchestration layers, static and dynamic analysis engines, and local CTI knowledge base integration
• Develop, design, and implement critical components across backend, infrastructure, and AI/ML layers
• Drive the development of agentic AI capabilities, leveraging LLMs, autonomous agents, and multi-agent orchestration frameworks (e.g., ReAct) for complex task execution in malware analysis workflows
• Integrate AI systems with traditional reverse engineering toolchains, including binary analysis, disassemblers, debuggers, and sandbox environments
• Lead LLM fine-tuning and adaptation on domain-specific datasets such as malware samples, CTI feeds, and behavioral telemetry
• Design and implement systems for air-gapped/on-prem deployments, including model lifecycle management, secure update mechanisms, data isolation, and system hardening
• Translate advanced research into production-grade systems, turning prototypes into scalable, reliable, and interpretable solutions
• Set and enforce a high technical standard, reviewing outputs such as YARA rules, feature extraction pipelines, model outputs, and system architecture decisions
• Ensure explainability and reliability of AI outputs, including generating human-readable analysis of complex behaviors and attack chains
  
  

• 5+ years of experience in software engineering, with proven ownership of architecture and system design for complex distributed systems
• Strong hands-on experience with low-level programming (C++ required, Rust is a strong advantage)
• Deep expertise in reverse engineering, including both manual and automated techniques, with experience in low-level systems analysis and feature extraction from binaries
• Proven experience working with LLMs, Deep Learning models, and ML pipelines, including fine-tuning, evaluation, and deployment in production environments
• Experience with multi-agent systems, autonomous agents, and orchestration frameworks (e.g., ReAct or similar paradigms)
• Strong background in ML engineering applied to security data, such as binary analysis, network traffic analysis, and CTI ingestion pipelines
• Experience designing systems for air-gapped or highly restricted environments, including secure infrastructure, isolation techniques, and offline model/data management
• Ability to operate across the full stack: from system architecture and workflow orchestration down to hands-on implementation and debugging
• High ownership, strong problem-solving capabilities, and the ability to drive ambiguous, complex challenges to completion
• Meticulous attention to detail, with a focus on accuracy, consistency, and interpretability of outputs