CAREERS / INTERNSHIPS
Build the future of AI security, trust, and governance tooling.
AI SECURITY ENGINEERING LABS COHORT
Labs cohort core
Structured, mentored, public-safe output
INTEGRATE
- MITRE ATLAS
- OWASP LLM Top 10
- NIST AI RMF
- ISO 42001 / AIMS
- EU AI Act
BUILD
- Labs tools
- Scanners & engines
- Evidence pipelines
- Scorecards
AI SECURITY ENGINEERING
Turn emerging AI security knowledge into useful systems.
RESEARCH
- AI risk
- Red / blue team
- Adversarial testing
- Trust governance
APPLY
- Partner orgs
- Operational feedback
- Case studies
- Continuous improvement
RESEARCH FIRST
Science, evidence, and transparent methodology.
REAL TOOLING
Build scanners, maps, scorecards, and workflows.
OPEN ECOSYSTEM
Integrate public frameworks, standards, and resources.
PORTFOLIO OUTPUT
Ship code, research notes, datasets, or published artifacts.
MENTORED & REVIEWED
Senior review, technical feedback, and scoped projects.
WHAT YOU'LL DO
Work at the intersection of AI, security, and governance.
Interns help build the aisecurity.llc Labs platform: scanners, scorecards, framework browsers, crosswalks, red-team scenarios, and governance evidence workflows. Work is scoped so each participant can ship a clear artifact by the end of the cohort.
BUILD & ENHANCE LABS TOOLING
- Improve framework browsers and navigators
- Build scanner outputs and scorecard views
- Add evidence checklists and public-safe caveats
- Improve UX for technical security content
INTEGRATE SECURITY FRAMEWORKS
- MITRE ATLAS and related attack mapping
- OWASP LLM Top 10 coverage
- NIST AI RMF and ISO 42001 / AIMS references
- Framework crosswalks and scorecard dimensions
RESEARCH & EXPERIMENT
- AI risk and safety patterns
- Red-team and blue-team scenarios
- Prompt injection and agentic misuse
- Telemetry, detection, and evaluation workflows
APPLY WITH PARTNER ORGS
- Turn operational feedback into tooling improvements
- Adapt Labs tools to real-world workflows
- Document sanitized case-study learnings
- Surface gaps between frameworks and practice
DOCUMENT & SHARE KNOWLEDGE
- Write research notes and implementation guides
- Publish sanitized findings and explainers
- Build examples, datasets, and crosswalks
- Help the ecosystem learn faster
WHO WE'RE LOOKING FOR
Curious builders who can learn in public and work with evidence.
You do not need to be a senior security engineer. You do need curiosity, discipline, strong communication, and the ability to turn vague questions into structured work.
- Students, researchers, recent graduates, or self-directed builders
- Interest in AI security, governance, product security, or risk
- Comfortable reading technical standards, repos, or documentation
- Able to write clearly about what you found and why it matters
- Willing to work asynchronously and document progress
- Comfortable receiving review and revising work
- Serious about evidence, caveats, and responsible claims
PROGRAM DETAILS
FORMAT
Remote, async-friendly
DURATION
10-16 weeks, cohort-based
TIME COMMITMENT
Approximately 5-10 hours per week unless otherwise agreed
COMPENSATION
Unpaid educational internship unless a paid posting says otherwise
MENTORSHIP
Structured project review, feedback, and office-hour style support
OUTPUT
Portfolio artifact, research note, code contribution, dataset, guide, or demo
WHAT INTERNS RECEIVE
- Defined project scope
- Mentorship and review
- Real-world AI security/governance context
- Portfolio-ready work
- Credit on published artifacts where appropriate
- Reference or recommendation where earned
- Exposure to frameworks, tools, and partner-informed workflows
- Potential consideration for future paid work if available
Important caveat
Participation does not guarantee employment, compensation, client access, or publication. Project access and attribution depend on quality, confidentiality, partner constraints, and review.
PROJECT TRACKS
Choose a track. Ship a concrete artifact.
Each track is scoped to help interns build a useful, public-safe artifact while learning the research and engineering patterns behind AI security work.
TRACK 1
FRAMEWORK INTELLIGENCE
Integrate and map AI security frameworks into usable Labs tools.
Example projects
- MITRE ATLAS compact navigator improvements
- OWASP LLM Top 10 browser
- NIST AI RMF crosswalk
- ISO 42001 / AIMS readiness model
Skills
TRACK 2
TRUST SCANNER & SCORECARDS
Turn public trust pages into evidence-backed scorecards and improvement guidance.
Example projects
- Trust-center artifact checklist
- AI policy language detector
- Public scorecard caveat system
- Security evidence backlog templates
Skills
TRACK 3
AI RED TEAM LABS
Build controlled educational scenarios for AI attack, misuse, and defense patterns.
Example projects
- Prompt injection scenario catalog
- Agentic misuse case studies
- Defensive pattern library
- Test harness examples
Skills
TRACK 4
BLUE TEAM & DETECTION
Translate AI security risks into telemetry, controls, and detection logic.
Example projects
- AI event taxonomy
- Logging and evidence models
- Detection engineering examples
- Governance telemetry mapping
Skills
TRACK 5
PRODUCT & UX FOR SECURITY TOOLS
Make complex AI security knowledge usable through clear interfaces.
Example projects
- Labs page improvements
- Matrix navigator density improvements
- Scorecard UX
- Documentation and demo flows
Skills
TRACK 6
RESEARCH WRITING & KNOWLEDGE BASE
Turn research into clear public knowledge assets.
Example projects
- Framework explainers
- Case-study summaries
- Tool comparison pages
- Guidance briefs
Skills
OPEN INTERNSHIP ROLES
Pick the role shape that matches how you want to contribute.
Each role is scoped around learning, mentorship, and a concrete portfolio artifact. The work may include research, tooling, writing, data modeling, UX, or implementation depending on track fit.
Internship role
AI Security Engineering Research Intern
Join a research-driven internship cohort building practical tools for AI security, trust, and governance. Interns contribute to Labs tooling, framework mapping, scorecards, and public-safe research artifacts.
Responsibilities
- Research AI security and governance frameworks
- Help map MITRE ATLAS, OWASP LLM Top 10, NIST AI RMF, and ISO 42001/AIMS themes
- Contribute to Labs tools, scorecards, or scanners
- Document findings and implementation notes
- Participate in async review and project check-ins
- Produce a portfolio-ready artifact
Good fit
- Curious about AI security, governance, or product security
- Comfortable reading technical docs and standards
- Clear writer and careful researcher
- Basic GitHub comfort
- Bonus: TypeScript, Python, Rust/WASM, Supabase, security testing, UX, or data modeling
Compensation
Unpaid educational internship unless otherwise stated.
Internship role
AI Trust Governance Intern
Help build tools that convert public trust, legal, security, and AI policy language into reviewable evidence, scorecards, and improvement guidance.
Responsibilities
- Review trust centers, privacy policies, terms, AI usage policies, and security pages
- Help define public trust artifact checklists
- Contribute to Trust Scanner rules, guidance, and scorecard mappings
- Research trust center patterns from leading AI and security organizations
- Draft public-safe guidance and caveats
- Build or improve knowledge-base content
Good fit
- Interest in AI governance, privacy, policy, legal tech, or security assurance
- Strong reading and writing skills
- Good judgment around claims and caveats
- Comfortable turning messy documents into structured data
Compensation
Unpaid educational internship unless otherwise stated.
Internship role
Labs Tooling Intern
Help improve the aisecurity.llc Labs platform: framework navigators, scanner pages, scorecard views, interactive matrices, and research tools.
Responsibilities
- Improve Labs pages and UX
- Build compact framework browsers
- Help integrate data from public repos and standards
- Create sample outputs and route demos
- Work with TypeScript/Next.js/Supabase where appropriate
- Document technical decisions
Good fit
- Frontend, full-stack, or data-oriented builder
- Interested in security tools and AI governance
- Comfortable with component systems and structured data
- Bonus: React/Next.js, Tailwind, Supabase, data visualization
Compensation
Unpaid educational internship unless otherwise stated.
COHORT MODEL
Small cohorts, scoped work, visible progress.
The internship is organized around short research-engineering cycles. Each intern selects or is assigned a scoped project, defines a deliverable, receives review, and ships an artifact by the end of the cohort.
Week 1
Orientation and project selection
Review the program, choose a track, confirm scope, and set success criteria.
Weeks 2-3
Research, source review, scope refinement
Map the problem, collect references, and define the artifact you will ship.
Weeks 4-8
Build, document, test, review
Implement the work, capture evidence, and iterate with mentor feedback.
Weeks 9-12
Polish, publish, demo, or hand off
Refine the deliverable and package it for portfolio use or final program handoff.
Weeks 13-16
Optional extension for deeper projects
Continue only if the cohort and project justify additional depth.
Exact timing can shift by cohort and project type.
REAL OPERATIONAL USAGE
The best tools are shaped by real workflows.
Where appropriate, interns may help adapt Labs tooling based on feedback from partner organizations, security teams, governance reviewers, or advisory projects. Partner-facing work is supervised and sanitized. Interns do not represent themselves as consultants or handle sensitive client material without explicit approval.
Boundary
Access to private partner data, client communications, and sensitive materials is limited and controlled.
Partner feedback
Turn reviewer questions, buyer objections, and governance gaps into clearer tooling requirements.
Workflow testing
Use sanitized scenarios to check whether Labs tools match real review and evidence flows.
Sanitized case studies
Document lessons from approved examples without exposing sensitive partner material.
Tool improvement backlog
Convert operational friction into scoped fixes, better UX, stronger rules, or clearer guidance.
OUTCOMES
By the end, you should have something real to show.
The internship is meant to leave you with a concrete artifact, a clearer understanding of AI security engineering, and a public-safe record of what you built.
A shipped Labs improvement
A framework integration or crosswalk
A scanner rule, scorecard field, or evidence model
A research note or technical guide
A demo, dataset, or public-safe artifact
A clearer understanding of AI security engineering as a discipline
APPLY
Apply for the next AI Security Engineering cohort.
Tell us what you can build, what you want to learn, and which track fits you best. We care more about evidence of curiosity, discipline, and clear thinking than credentials.
APPLICATION CHECKLIST
A strong application can be short. Show us how you think, what you have tried, and what you want to build.
Expected submission shape
Send a project idea, a portfolio link, or a short note about what you want to learn. Strong applicants are matched to scoped research and engineering tracks.
What happens next
- 1Send a short note and the link(s) you want us to review.
- 2We review fit, availability, and track alignment.
- 3Short follow-up if we need scope clarification or timing details.
- 4If accepted, we assign a scoped project and cohort start window.
FAQ
Frequently asked questions
READY TO BUILD
Ready to build useful AI security tooling?
Apply with a project idea, a portfolio link, or a short note about what you want to learn. We’ll match strong applicants to scoped research and engineering tracks.