AI Security Researcher, Harmonic Security
+ Equity
Zero-touch data protection
Be an early applicant
Zero-touch data protection
21-100 employees
Be an early applicant
+ Equity
21-100 employees
Company mission
To pave the way for secure AI adoption in enterprises, ensuring innovation thrives without compromise.
Role
Who you are
- Deep Technical Curiosity: 3-5 years of experience in cybersecurity or applied ML/AI security research, with a strong interest in how AI systems fail and can be attacked
- AI Security Expertise: Hands-on experience assessing or red-teaming LLMs, generative AI tools, or AI-integrated SaaS products; familiarity with adversarial ML, model extraction, prompt injection, or data poisoning
- Security Research Skills: Proven ability to design and execute independent research- identifying vulnerabilities, publishing technical findings, and proposing mitigations
- Communication & Storytelling: Excellent written and verbal skills for translating complex research into accessible content - whether blog posts, practitioner guides, or conference talks
- Public Presence: A strong desire to build a profile for presenting or contributing to security or AI conferences (e.g., Black Hat, DEF CON, RSA)
- You’re comfortable figuring things out from scratch and don’t wait for detailed instructions to start making progress
- You’re naturally curious - you like to dig into how AI systems actually work and break, and you keep up with new developments just because you find them interesting
- You like solving hard, undefined problems and can turn ideas into experiments or results quickly
- You enjoy working with others, sharing what you learn, and helping teammates succeed
- You take pride in clear, honest communication - whether explaining a technical finding or giving feedback
- You get energy from teaching others, writing, or speaking publicly about your work
- You care about impact more than titles or process - you just want to enable other to use AI securely
What the job involves
- Harmonic’s Security & TechOps team owns the systems, tooling, and research that keep Harmonic secure, trusted, and moving fast
- We work across the entire lifecycle of how the company operates-from how we build software, to how we detect threats, to how we prove our security posture to customers, to how employees use the technology that powers their work
- We focus on enabling speed without sacrificing safety
- That means creating secure-by-default engineering practices, building automation and detections that reduce manual work, researching emerging AI security risks, running assurance programmes like SOC 2 and ISO 27001, and shaping the internal platforms that help teams work efficiently and securely
- We’re a hands-on, engineering-first group
- We avoid box-ticking and heavy processes in favour of practical solutions, strong automation, and constant improvement
- We collaborate closely across functions, operate with high ownership, and support each other through rapid growth
- Our shared goal is to build the trust, systems, and capabilities that let Harmonic innovate confidently and at scale
- Harmonic’s Security team is breaking new ground in understanding the security risks that come with Generative and Agentic AI
- Our mission is to equip security leaders and practitioners with clear, actionable insights on securing AI systems
- This work should shape how the industry thinks about AI security, and our findings make their way onto the stages of the world’s top security and AI conferences
- As we embark on this mission, we’re looking for a founding AI Security Researcher to build and lead Harmonic’s AI Security Research program from the ground up
- You’ll define our research agenda, uncover real-world security risks in AI tools and platforms, and help shape how enterprises can adopt AI safely
- This role blends hands-on technical research with public thought leadership - from identifying vulnerabilities in emerging AI systems to creating guides, publishing insights, and representing Harmonic at major industry events
- You’ll also play a key role in informing our product strategy, ensuring Harmonic stays ahead of evolving AI security threats and opportunities
- Lead AI Security Research: Design and drive a greenfield research program focused on securing the use of AI systems and tools
- Thought Leadership: Develop practitioner guides and actionable content on securing AI technologies (e.g., “How to secure ChatGPT Enterprise”)
- Public Engagement: Represent Harmonic at industry conferences (e.g., RSA, BlackHat, DEFCON) through talks, panels, and research presentations
- Threat Intelligence: Research emerging AI threats and threat actors, assessing their impact on our customers and the broader ecosystem
- Internal SME: Partner with Product Security to identify and mitigate AI-related risks across Harmonic’s products and infrastructure
- What Success Looks Like (6–12 months)
- Design, build, and operate a dedicated AI security research environment, enabling safe testing of generative and agentic AI systems and reproducible research workflows
- Publish and present new, original research at a major security or AI conference (e.g., Black Hat USA, DEF CON, BSides, or equivalent) that offers actionable insights for practitioners and security leaders
- Establish a consistent content pipeline, producing high-quality research write-ups, blogs, demos, or practitioner guides that clearly explain risks, attack paths, and defensive strategies
- Develop recognisable external presence, securing initial press coverage or analyst mentions that highlight Harmonic’s leadership in AI security
- Contribute research that shapes internal product direction, informing detection capabilities, product security priorities, or policy decisions based on real-world risk findings
Our take
The power of generative AI has made businesses across industries move quickly to adopt the technology. But this quick adoption has raised security concerns, especially surrounding how businesses can protect sensitive information without slowing down innovation.
This is where Harmonic Security steps in, with data protection built for the gen-AI era. It tracks AI adoption, use cases and risk, helping businesses to gain visibility into how their employees are using gen-AI, and monitor the exposure of their data to risky applications.
This approach means that businesses can detect data privacy risks while adopting gen-AI safely. So it's a welcome tool for those who want to build out a competitive edge without risking data breaches. It now plans to scale up its engineering and go-to-market efforts off the back of a successful funding round.
Kirsty
Company Specialist at Welcome to the Jungle
Company
Funding (2 rounds)
Oct 2024
$17.5m
SERIES A
Oct 2023
$7m
SEED
Total funding: $24.5m
Company HQ
Civic Center, San Francisco, CA
Leadership
Alastair Paterson
(Co-Founder & CEO)Board Member at GBx. Previously CEO at Digital Shadows, International Propositions Manager at Detica and worked in Mobile/Internet Research and Development at Hewlett-Packard.
Bryan Woolgar-O'Neil
(Co-Founder & CTO)Previously VP of Engineering at ReliaQuest, CTO at Digital Shadows, Technical Architect/Lead Developer at Aimia and Senior Developer at Detica.
Share this job
View 6 more jobs at Harmonic Security