Senior Software Engineer | AI Security Engineering

About the Role

NinjaOne’s Application Security Engineering team is looking for a Senior Software Engineer to assist in building an AI platform that discovers vulnerabilities in production code, validates findings against real attack paths, and proposes tested fixes. You will build and own the software that makes this work reliable at scale, partnering with Security Architecture and Product Engineering to turn security research into actionable improvements across our cloud platform and endpoint agents.

This is a hands-on software engineering role with ownership from design through production. You will solve problems in agent orchestration, durable execution, context management, evaluation, and cost control. Your work will help engineers find and fix meaningful vulnerabilities earlier, with evidence they can trust.

We are hiring four engineers across two focus areas: Vulnerability Discovery and Finding Validation, with two openings in each. We will discuss your preference during interviews. Discovery develops and tests vulnerability hypotheses; Validation independently challenges findings and develops evidence-backed fixes. Separate model families and independent ownership help each stage check the other’s conclusions.

Location - We are flexible on remote working from home, if you are located in the USA and reside in one of the following states - CA, CO, CT, FL, GA, *IL, KS, MA,MD, ME,NJ, NC, NY,OH, OR, TN, TX, VA, and WA. We have physical offices in Austin, TX and Tampa, FL, if you prefer a hybrid option.

• Onsite interviews may be required for this role.

We hire the best software engineers, but experience in our stack can’t hurt: NinjaOne is built on Java, Kotlin, C++, Golang and Postgres; supporting millions of user endpoints and running as a scalable cloud service in AWS. Knowing large-scale datastore bottlenecks, asynchronous application design and client-server architecture will help you.

What You’ll be Doing

• Build and operate a production AI vulnerability platform with resumable workflows, bounded agent execution, model routing, and per-repository budgets across AWS Bedrock and OpenAI-compatible APIs.
• Engineer repository context, threat models, and coverage tracking so agents can reason about entry points, trust boundaries, tenant isolation, and relevant code paths.
• Build independent validation, stable finding identity, and deduplication. Ground reachability and exploitability of judgments in code, deployment evidence, and reproducible tests.
• Develop patch and regression-test workflows that demonstrate a failing test before a fix and a passing test afterward, check for regressions, and require human review before merging.
• Protect agent execution with sandboxing, least-privilege tool access, and controls for untrusted repository content, prompt injection, secrets, and sensitive data.
• Partner with security and engineering teams to deliver findings through SARIF and existing workflows. Measure validated finding quality, coverage, reliability, and cost per validated finding; communicate results and share engineering practices.
• Other duties as needed

About You

• 5+ years of software engineering experience, including ownership of production services or automation platforms, and hands-on experience building LLM-backed applications or agent workflows.
• Strong Python skills, including asynchronous concurrency, typing, testing, packaging, and profiling. You can design, ship, debug, and maintain production-quality software.
• Experience building multi-step workflows with durable state, idempotent retries, task queues, termination conditions, observability, and resource limits. You can explain the failure modes and tradeoffs of systems you have shipped.
• Hands-on experience with LLM APIs, tool calling, structured outputs, context management, and evaluation. You can assess when to use an agent framework and when simpler orchestration is more reliable.
• Experience with AWS or comparable cloud infrastructure and distributed systems. Familiarity with model provider integrations, container isolation, and secure handling of code and credentials.
• Working knowledge of application security, including authentication, authorization, common vulnerability classes, and trust boundaries. You can use evidence to distinguish a plausible hypothesis from an actionable finding.
• Ability…