S
Senior Backend Engineer
Sin especificarTiempo completoBackend DevelopmentCybersecurity
Build production security systems that turn threat research into defenses for enterprise digital assets. Work with product, security research, and AI teams on backend and full-stack capabilities.
Responsabilidades
- Turn attack research into production defenses that discover exposed assets, identify risky access relationships, and interrupt account-takeover paths.
- Build browser automation and reverse engineer platform APIs to uncover permissions, reconstruct account state, and execute protection workflows.
- Design distributed systems that correlate account activity, identity changes, and threat signals into findings and coordinated responses.
- Develop AI investigation workflows using language models, contextual retrieval, and agents with permission-bound tools and independently verified actions.
- Drive security features from architecture to production, combining investigation interfaces with resilient backend services, tenant isolation, and audit trails.
Requisitos
- At least 8 years of backend or full-stack development experience, including ownership of production systems.
- Engineering experience at a cybersecurity product company.
- Strong TypeScript and Python skills and experience building modern frontend applications.
- Experience with browser automation and reverse engineering APIs, including authentication, sessions, tokens, and undocumented endpoints.
- Hands-on experience with RabbitMQ, Redis, Elasticsearch, and Milvus.
- Strong foundations in distributed systems, secure API design, automated testing, CI/CD, and production observability.
- Experience integrating LLM APIs, structured outputs, embeddings, retrieval-augmented generation, and tool calling into application workflows.
- Practical experience with AI development tools, supported by independent code review, testing, and security validation.
- Experience evaluating AI workflows with representative datasets, regression tests, and tracing, considering accuracy, latency, and cost.
- Understanding of prompt injection, sensitive-data exposure, and agent permissions, with authorization enforced independently of model output.