Software Engineer – AI-Enabled Vulnerability Research (SWE3) [D.26.0226]

Applicants must already hold a TS/SCI clearance with a current Full Scope Polygraph before applying. We are not sponsoring or processing clearances for this position. Candidates who are seeking to obtain a clearance do not meet the minimum requirements.

Salary Range: $260k-$285k per year with an additional $65k-$71k in immediately vested company 401(k) contributions

Description: We are seeking an experienced Software Engineer to design, develop, and integrate AI-enabled capabilities supporting computer network exploitation (CNE) and vulnerability research (VR). This role combines strong software engineering fundamentals with practical knowledge of vulnerability research, program analysis, and AI-assisted research workflows. We are particularly interested in candidates with hands-on experience conducting dynamic vulnerability research in isolated sandbox environments. You will independently own complex features and subsystems, translate mission needs into technical designs, and deliver reliable software from prototype through deployment. You will develop agentic AI harnesses that connect large language models, research tools, instrumented execution environments, and backend services to support repeatable, evidence-driven analysis. Working with vulnerability researchers, AI/ML engineers, and mission stakeholders, you will make sound architectural decisions, evaluate AI-generated findings, and improve the quality and effectiveness of research workflows. This position also includes mentoring engineers and contributing to technical direction and engineering standards.

Responsibilities:

• Design, develop, and maintain AI-enabled software and research tooling supporting CNE and vulnerability research workflows.
• Own the design and delivery of complex features and subsystems, including requirements refinement, implementation, testing, deployment, and sustainment.
• Develop and integrate agentic AI harnesses that orchestrate research tools, manage execution state, collect evidence, and evaluate results.
• Build and extend sandbox-based dynamic analysis capabilities for executing and observing software, capturing runtime behavior, and reproducing suspected vulnerabilities.
• Integrate debugging, instrumentation, fuzzing, crash analysis, and related research tools into automated or researcher-directed workflows.
• Establish repeatable evaluations that measure finding validity, reproducibility, false-positive rates, and the usefulness of AI-assisted analysis.
• Develop backend services, APIs, data pipelines, and user interfaces that make research capabilities accessible and results traceable.
• Troubleshoot complex issues across application code, AI integrations, research tooling, and execution environments; improve performance, reliability, and observability.
• Lead design and code reviews within assigned areas, document technical decisions, mentor engineers, and communicate progress, risks, and tradeoffs to stakeholders.

Skills Requirements:

• Demonstrated experience independently designing, implementing, testing, and delivering complex software features or subsystems, with sound judgment about architecture, maintainability, and operational requirements.
• Proficiency in Python and experience building maintainable, tested software using modular design and appropriate object-oriented programming practices.
• Demonstrated expertise in AI-enabled CNE or vulnerability research, including integrating AI capabilities into research workflows and critically evaluating their outputs.
• Experience with vulnerability research or software analysis techniques, such as reverse engineering, debugging, dynamic analysis, fuzzing, or crash triage.
• Working knowledge of isolated execution environments and dynamic analysis concepts, including runtime instrumentation, execution tracing, environment reproducibility, and evidence collection.
• Experience developing LLM applications or agentic systems that use external tools, structured outputs, workflow state, error handling, and evaluation.
• Experience developing backend services and REST APIs using frameworks such as FastAPI, Flask, or Django, and working with relational databases such as PostgreSQL.
• Experience working in Linux environments, with an understanding of operating system behavior, processes, memory, and networking relevant to software analysis.
• Experience using Git, automated testing, CI/CD, and code review to deliver reliable software collaboratively.
• Ability to translate ambiguous requirements into actionable technical work, explain technical tradeoffs, and provide effective guidance to other engineers.

Nice to Haves:

• Hands-on experience conducting dynamic vulnerability research in sandbox environments, including configuring instrumented targets, controlling execution, capturing runtime evidence, and reproducing findings across clean environment resets.
• Experience building AI agents that interact with debuggers, fuzzers, emulators, or other analysis tools within controlled research environments.
• Experience with virtualization, emulation, snapshots, and automated provisioning of repeatable analysis environments.
• Experience with C/C++, assembly, binary analysis, memory corruption analysis, or operating system internals.
• Experience designing evaluation datasets and benchmarks for AI-assisted vulnerability discovery, triage, or validation.
• Experience with retrieval-augmented generation, vector databases, or search platforms such as Elasticsearch, Qdrant, or ChromaDB.
• Experience building researcher-facing interfaces using TypeScript, React, or comparable technologies.
• Contributions to vulnerability research, security tooling, open-source projects, or relevant technical publications.

YOE Requirement: 12 yrs., B.S. in a technical discipline or 4 additional yrs. in place of B.S.