Website:
cosmipher.com
Job details:
Company Description Cosmipher Tech Labs secures AI systems that take actions, including models, agents, MCP servers, and the identities, tools, and data behind them. The company delivers a unified boundary of products across discovery, posture management, runtime control, adversarial security testing, and AI supply chain integrity, with additional focus on synthetic media through DeepfakeWatch. Cosmipher emphasizes rigorous, transparent security practices, starting every engagement with written scope, lawful authority, exclusions, stop conditions, and accountable ownership of findings. The team builds for security leaders, platform teams, and governance professionals who need verifiable evidence of AI risk and control, not just assurances. Cosmipher helps organizations make high-impact decisions about AI estates, agent actions, and release candidates in a disciplined, security-first manner.
Role Description The AI Security Developer is a full-time, remote role focused on designing, implementing, and hardening security controls for AI systems, agents, and supporting infrastructure. In this role, the developer will build and maintain secure software components for products such as AI firewalls, security testing frameworks, and supply chain integrity tools, using modern programming practices and security principles. Daily responsibilities include analyzing AI behavior and patterns, developing detection and enforcement logic, integrating security features into AI pipelines, and collaborating with security researchers to translate adversarial findings into resilient code. The developer will work closely with product, platform, and governance teams to define technical requirements, review architectures, and contribute to documentation and evidence that support compliance, audits, and security assessments. The role also involves continuous improvement of AI security tooling, participating in code reviews, and staying current with emerging threats and techniques in AI and cybersecurity.
Qualifications
- Strong foundation in Computer Science and Software Development, including data structures, algorithms, secure coding practices, and experience with one or more modern programming languages (e.g., Python, Go, Rust, or similar).
- Skills in Neural Networks and Natural Language Processing (NLP), with experience building, evaluating, or securing ML models or AI agents in production environments.
- Pattern Recognition expertise applied to detecting anomalous or adversarial behaviors in AI systems, logs, and agent activity across complex environments.
- Understanding of AI security concepts, including adversarial ML, AI supply chain risks, identity and access control for agents, and runtime protection mechanisms.
- Familiarity with security engineering or application security (e.g., threat modeling, code review, penetration testing, or secure architecture design) is highly beneficial.
- Ability to work independently in a remote environment, collaborate across functional teams, and communicate technical findings clearly in writing.
- Bachelor’s degree in Computer Science, Engineering, or a related field, or equivalent practical experience in
Click on Apply to know more.