Volunteer AI Safety Research Collaborator

ABOUT THE OPPORTUNITY
ZeroBug Technologies is building a small independent research group focused on AI safety, LLM security, adversarial robustness and identifying vulnerabilities in AI systems.
We are currently bootstrapped and have no external funding, so this is an UNPAID, VOLUNTARY RESEARCH COLLABORATION. It is not a salaried position and there is no guarantee of future employment or compensation.
We want to work with people who are genuinely interested in understanding how AI systems fail and how they can be made safer.
WHAT WE WILL RESEARCH
Our work may include:
• AI safety and LLM security
• Adversarial robustness and red teaming
• Jailbreak and safeguard evaluation
• Prompt injection and agentic AI security
• Hallucinations and information reliability
• Safety evaluation methodologies
• Responsible vulnerability research
• Designing and evaluating potential mitigations
The focus is not simply finding ways to break AI systems. Research should help us understand why failures occur, measure them responsibly and explore how systems can be made safer.
HOW IT WILL WORK
Researchers will collaborate remotely on selected research questions and experiments. We will document the methodology, results and limitations and aim to turn strong projects into public research.
Where research reaches publication quality, ZeroBug Technologies will support the process of preparing the work for appropriate journals, workshops, conferences or other research venues.
Contributions will be credited appropriately. Authorship on research papers will be based on actual research contribution and standard academic authorship principles.
WHO CAN APPLY
You do not need to already be an AI safety researcher.
We are interested in students, graduates, engineers, data scientists, cybersecurity researchers and independent researchers who are curious about AI safety and willing to learn.
Useful experience includes Python, machine learning, LLMs, cybersecurity, data analysis, academic research or technical writing, but strong curiosity and research thinking are equally important.
IMPORTANT: COMPENSATION
This collaboration is currently unpaid.
ZeroBug Technologies is bootstrapped and currently has no funding allocated for researcher compensation. We want to be completely transparent about this before anyone applies.
There is no promise or guarantee of future payment. If the company later obtains funding that allows us to create paid research positions, our intention is to transition active roles to paid positions where possible, subject to funding, role requirements and applicable agreements.
There are no application fees or participation fees.
WHY WE ARE DOING THIS
AI systems are becoming increasingly capable and increasingly integrated into everyday life. We want to build a research community that independently investigates where these systems fail, documents those failures responsibly and contributes useful evidence toward making AI systems safer.
If that sounds like something you would genuinely enjoy researching, we'd like to hear from you.

Work arrangement
No