AI Security Laboratory: Hands-On + Full-Stack (Lifetime Lab Access) - Dawid Czagan - DCTSG2027
Description
Name of Training: AI Security Laboratory: Hands-On + Full-Stack (Lifetime Lab Access): Exploit and Defend LLM Applications, Build Private AI Workflows and Agentic AI, and Harness AI for Security Research and Everyday Cybersecurity Tasks
Trainer(s): Dawid Czagan
Skill Level: Intermediate
Dates: March 5 – March 6, 2027
Time: 9 am to 5 pm
Venue: TBD
Early Bird Cost: S$4400 (incl. GST, until Dec 31, 2026)
Regular Cost: S$4700 (incl. GST)
Short Summary
Master AI security from both attacker and builder perspectives with intensive, hands-on experience. Attack and fuzz LLM applications, build AI-powered security workflows and agentic AI, and apply AI to real-world security operations, vulnerability research, and PoC development, with reusable Python scripts and lifetime lab access for continued practice.
Course Description
AI has introduced an entirely new layer of security risk — one that needs to be understood from both attacker and builder perspectives. This training is a hands-on, full-stack guide to that landscape, showing how modern AI systems are attacked, built, and used in real-world security.
You will work through the offensive side of AI security with prompt injection, jailbreaking, and fuzzing of LLM applications. You will also apply AI in security engineering and daily operations — turn AI building blocks into practical workflows and use AI security tools for everyday tasks.
Along the way, you will move into more advanced capabilities — building agentic AI for real-time security operations, applying AI to vulnerability research and PoC development, and exploring how smarter AI can assess other AI.
The training includes hands-on exercises, reusable Python scripts, and lifetime lab access — so you can continue practicing and applying what you learned long after the class ends.
Key Learning Objectives
- prompt injection: direct and indirect
- LLM jailbreaking
- fuzzing LLM applications
- AI-powered shell
- advanced prompting
- local LLMs / private AI
- AI programming
- AI attack detection
- OpenAI models and API
- embeddings
- quantization
- LLM Guard
- building agentic AI
- CVE research / PoC development with AI
- smarter AI assessing other AI
- specialized AI security tools
- and more …
What Students Say About My Trainings
Recommendations are available on my LinkedIn profile (https://www.linkedin.com/in/dawid-czagan-85ba3666/) – training participants from companies such as Oracle, Adobe, ESET, ING, Red Hat, Trend Micro, Philips, the government sector, and others.
Course Outline
View full outline
Hide outline
Day 1 - Morning
- AI/LLM building blocks — exploring tokens, context window, prompt types, and non-deterministic behavior, with hands-on playground exercises
- High-impact AI attacks — prompt injection / RAG poisoning, jailbreaking, prompt leakage, and adversarial attacks
- Using AI in security practitioners’ daily operations — including AI-powered shell and advanced prompting
- Smarter AI assessing other AI — along with real-world security applications
Day 1 - Afternoon
- AI programming for security practitioners — working with local LLMs / private AI and cloud models (OpenAI / API)
- Building AI workflows for security use cases (e.g. fully private AI setups)
- Ready-to-use Python scripts, providing hands-on experience and reusable AI building blocks for daily security tasks
- Prompt injection attacks in depth — direct and indirect prompt injection, creative attack techniques, and multimodal attack vectors
Day 2 - Morning
- Fuzzing LLM applications, which differs from traditional fuzzing due to the non-deterministic nature of modern LLMs
- AI attack detection — using both cloud and local LLMs to identify malicious inputs and toxic content
- Anonymizing data before sending it to cloud LLM providers
- Applying open-source defenses such as LLM Guard
- Specialized AI security tools
Day 2 - Afternoon
- CVE research and PoC development with AI
- Building agentic AI for real-time security operations
- Interesting AI security projects
- Q&A session
- Proficiency evaluation
Difficulty Level
Intermediate - The student has education and some experience in the field and familiarity with the topic being presented. The student has foundational knowledge that the course will leverage to provide practical skills on the topic.
Suggested Prerequisites
Students should have a general understanding of application security and some experience with web technologies and APIs. Basic familiarity with programming or scripting, security testing practices, and working with the command line is recommended.
What Students Should Bring
Students will need a laptop with 64-bit operating system, at least 16 GB RAM, 120 GB free hard drive space, administrative access, ability to turn off AV/firewall and VMware Player/Fusion installed (64-bit version). Prior to the training, make sure there are no problems with running x86_64 VMs. You will need an OpenAI API key (required). A Lakera API key is optional.
As AI is an evolving field, additional requirements may be shared ahead of the training if needed.
What the Trainer Will Provide
Students will be given a VMware image with a specially prepared lab environment to work on many topics and exercises in this training. When the training is over, students can take the lab environment home (after signing a non-disclosure agreement) to continue practicing at their own pace.
Trainer(s) Bio
Dawid Czagan is an internationally recognized security researcher and trainer. He is listed among top hackers at HackerOne. Dawid Czagan has found security bugs at Apple, Google, Mozilla, Microsoft and many others. Due to the severity of his findings, he received numerous awards.
Dawid Czagan shares his offensive security experience in his hands-on training courses. He delivered training sessions at key industry conferences such as DEF CON (Las Vegas), OWASP Global AppSec EU (Barcelona), Hack In The Box (Amsterdam), CanSecWest (Vancouver), 44CON (London), Hack In Paris (Paris), NorthSec (Montreal), HITB GSEC (Singapore) and for many corporate clients. His students include security specialists from Oracle, Adobe, ESET, ING, Red Hat, Trend Micro, Philips and the government sector. Recommendations are available on Dawid Czagan's LinkedIn profile (https://www.linkedin.com/in/dawid-czagan-85ba3666/).
Dawid Czagan is the founder and CEO at Silesia Security Lab. To find out about the latest in his work, you are invited to subscribe to his newsletter (https://silesiasecuritylab.com/newsletter) and follow him on Twitter (@dawidczagan), YouTube (https://www.youtube.com/channel/UCG-sIlaM1xXmetFtEfqtOqg), and LinkedIn (https://www.linkedin.com/in/dawid-czagan-85ba3666/).
Proficiency Exam
After completing the course, students will take a real-world, scenario-based evaluation that assesses practical skills. Students may use all course materials and notes during the evaluation. A minimum score of 65% is required to successfully demonstrate proficiency. Each student will receive feedback on their performance.
Registration Terms and Conditions
Trainings are refundable before January 29, 2027, minus a non-refundable processing fee of $350.
Between January 30, 2027 and February 28, 2027 partial refunds will be granted, equal to 50% of the course fee minus a processing fee of $350.
All trainings are non-refundable after Feb 28, 2027.
Training tickets may be transferred to another student. Please email us at training@defcon.org for specifics.
If a training does not reach the minimum registration requirement, it may be cancelled. In the event the training you choose is cancelled, you will be provided the option of receiving a full refund or transferring to another training (subject to availability).
Failure to attend the training without prior written notification will be considered a no-show. No refund will be given.
DEF CON Training may share student contact information, including names and emails, with the course instructor(s) to facilitate sharing of pre-work and course instructions. Instructors are required to safeguard this information and provide appropriate protection so that it is kept private. Instructors may not use student information outside the delivery of this course without the permission of the student.
By purchasing this ticket you agree to abide by the DEF CON Training Code of Conduct and the registration terms and conditions listed above.
Several breaks will be included throughout the day. Please note that food is not included.
All courses come with a certificate, contingent on attendance at 75% of course sessions. Students who pass the end-of-course evaluation will be issued a certificate of proficiency. All other students with sufficient course attendance will be issued a certificate of completion. Students are responsible for making sure their attendance is recorded at the start of the day and after returning from lunch.
