SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Coins is the most established crypto brand in the Philippines, serving over 18 million users through an easy-to-use mobile app for buying, selling, and trading cryptocurrencies and financial services. The company is fully regulated by the Bangko Sentral ng Pilipinas (BSP) and holds both Virtual Currency and Electronic Money Issuer licenses—the first crypto company in Asia to achieve this dual certification.
You will join as an AI Security Engineer responsible for securing the company's AI Agent infrastructure against emerging threats. Your work spans both offensive and defensive security domains.
Key responsibilities include:
- Conduct AI Agent security reviews and penetration testing, designing and implementing defenses against AI tool misuse, context poisoning, data poisoning, and prompt injection attacks
- Design and develop AI-based automated security detection tools to enable alert triage, attack attribution, and improved security operations efficiency
- Build Agent behavior auditing and anomaly detection systems to identify abnormal behavior chains during runtime
- Construct an AI-driven security operations platform and implement defense-in-depth solutions based on real business scenarios
- Assess attack surfaces of Agent runtime environments by combining security risks of cloud-native infrastructure (Kubernetes, Docker, cloud services) and design comprehensive defenses
- Work cross-functionally with other teams to implement security improvements that enhance customer experience
This is a high-impact role in a fast-growing organization expanding beyond APAC, offering meaningful career advancement opportunities.
REQUIREMENTS:
- Minimum 3+ years of security offense/defense experience (penetration testing, red team practice), with at least 1+ year in LLM (large language model) security offense/defense; extensive hands-on red team experience preferred
- Familiarity with network architecture and cloud-native technology stacks including Kubernetes, Docker container security, and cloud security (IAM, network isolation, cloud-native threat models); ability to assess security risks in complex infrastructure
- Practical Agent development capability: ability to independently build/modify AI Agents using frameworks such as LangChain, and hands-on offensive capability against Agents (construction and validation of attack techniques including tool misuse, prompt injection, context poisoning)
- Familiarity with security frameworks such as OWASP Top 10 for LLM and MITRE ATT&CK; knowledge of common AI attack techniques, adversarial logic, and defense solutions; strong passion for AI security
- Proficiency in at least one programming language (Python, Go, or C++); ability to develop security offense/defense tools or perform deep secondary development on open-source security platforms
Nice to have:
- Familiarity with mainstream AI frameworks (PyTorch, TensorFlow, LangChain); practical experience in local deployment and optimization of large language models
- Hands-on cloud security (AWS) experience or related certifications
