AI Red Teaming Project Ideas
Defending Systems: Configuring Firewalls, Network Monitoring, and Log Analysis
Apply for a Seat in the Lab
Project Implementation Portfolios
Browse our structured blueprints and developer checklists for building enterprise AI Red Teaming Training systems. Mapped directly to CACTS 1-to-1 mentoring.
Key Takeaways
- Use automated probing frameworks like PyRIT and NVIDIA Garak to test and analyze LLM prompt interaction.
- Set up open-source LLM Moderation Middleware platforms to aggregate and search server security logs.
- Configure firewalls and network access controls to enforce defensive architectures.
Project 1: Direct & Indirect Prompt Injection Audit
Execute direct and indirect prompt injection attacks to bypass LLM system instructions. Identify unencrypted HTTP requests, trace DNS queries, and detect unauthorized LLM vulnerability probing actions. Write a report detailing how to secure the network against these vulnerabilities.
Project 2: Automated AI Red Teaming Scans with PyRIT & Garak
Configure a secure Linux (Ubuntu) server. Disable unused services, configure UFW (Uncomplicated Firewall) rules, change default SSH ports, and implement SSH key-based authentication. Set up fail2ban to block IP addresses with repeated failed login attempts.
Project 3: GenAI Moderation Middleware & NeMo Guardrails Setup
Install an open-source LLM Moderation Middleware tool (like Elastic Stack/Wazuh). Direct system logs, authentication events, and firewall alerts from separate virtual machines to the LLM Moderation Middleware dashboard, and write alerts to trigger on unauthorized root logins.
Project & Syllabus FAQs
When does the next 1-to-1 training intake start?
Intakes start twice monthly on the 1st and 15th. The next upcoming 1-to-1 intake starts on October 1, 2026 (with secondary intake on October 15, 2026).
What real production projects will I build during the AI Red Teaming lab?
You will build end-to-end production systems using PyRIT, OWASP LLM Top 10, NVIDIA Garak, NeMo Guardrails, Burp Suite, NeMo Guardrails, Linux, starting from architecture blueprints to live staging server deployments.
Are student projects deployed on live cloud servers with public links?
Yes. Every project is committed to your personal public GitHub portfolio and deployed to live staging environments so recruiters can verify your working application.
How do portfolio projects help in clearing technical screening interviews?
Live project links and clean GitHub commit graphs serve as undeniable proof of engineering capability, allowing you to walk interviewers through actual production code during technical discussions.
Student Success Stories
Real feedback from students who completed our 1-to-1 virtual AI Red Teaming Training training.
"The AI Red Teaming lab at CACTS is outstanding. The 1-to-1 virtual session format allowed me to execute direct and indirect prompt injection attacks, bypass LLM guardrails like NeMo, and test OWASP Top 10 for LLM vulnerabilities on active staging applications."
Akash P.
AI Security Analyst, Kharadi, Pune"Learned offensive AI security, prompt injection payloads, and RAG data exfiltration defense in a practical 1-to-1 environment. My mentor demonstrated how to hijack multi-agent workflows and build custom moderation guardrails line-by-line."
Tanvi S.
LLM Security Researcher, Baner, Pune"I needed hands-on experience auditing GenAI applications and AI agents. We used Microsoft PyRIT and NVIDIA Garak to automate vulnerability scans and build defense middleware. The trainer's practical AI security experience was evident throughout."
Gaurav H.
Application Security Audit Specialist, Kothrud, PuneProduction Staging & Git Code Review Protocol
How students compile, test, and deploy production software code during the AI Red Teaming & Security track.
4-Stage Software Engineering Pipeline
- Feature Branch Isolation: You create dedicated Git branches for each module component following industry GitFlow standards.
- Live Screen Code Review: Your mentor reviews pull requests line-by-line, pointing out security flaws and performance optimizations.
- Automated Testing: You run unit and integration test suites before merging into staging.
- Staging Deployment: Deploy active builds onto cloud servers or live hosting environments.
Verified Portfolio Showcase
Employers don't want to see simple copy-paste tutorial code. Your completed projects become part of your public GitHub portfolio, backed by your verified completion credentials on verify.html and real experience in our Live Project Internship Program.