We are looking to hire a Head of Engineering and Architecture AI Securitywho can lead the design and development of our AI security capabilities while actively contributing to implementation. Please prioritize candidates with strong practical expertise in AI and cybersecurity, supported by hands-on engineering and architecture experience.Experience and background

- Demonstrated AI security experience: Practical experience identifying vulnerabilities in AI/ML systems, evaluating their resistance to attacks, and implementing security controls for real applications.

- Security product development: Proven experience designing, building and operating security products or platforms, with direct ownership of detection logic, security architecture, evaluation methods and implementation.

- Strong AI/ML understanding: Experience working with language models, model inference, classification, embeddings and evaluation pipelines, including an understanding of model limitations and failure modes.

- Recent hands-on implementation: Ability to write and review code, integrate models and security components, investigate vulnerabilities, and resolve performance or reliability issues.

- Technical leadership: Experience guiding AI/ML and security engineers, making architecture decisions, and translating security requirements into working product capabilities.

Core technical requirements

- LLM and AI application security: Understanding of prompt injection, jailbreaks, sensitive-data leakage, insecure model outputs and misuse of AI tools.

Experience securing prompts, retrieved content, model responses and agent workflows.





- Adversarial testing and security evaluation: Ability to design attack scenarios, build repeatable test suites, assess vulnerabilities and validate mitigations. Understand false positives, missed attacks, test coverage and the limitations of LLM-based evaluators.

- AI/ML engineering: Practical knowledge of model inference, text classification, embeddings, training and evaluation data, confidence calibration and threshold selection.

Experience with PyTorch, scikit-learn, Hugging Face, sentence-transformers, SetFit or comparable frameworks.

- AI data and model integrity: Understanding of data provenance, model lineage, artifact versioning, cryptographic hashing and signing, and risks associated with untrusted datasets, dependencies and model artifacts.

- Security architecture: Strong knowledge of threat modeling, authentication, authorization, tenant isolation, encryption, secrets management and audit logging. Ability to establish clear trust boundaries around models, applications, tools and customer data.

- Python and AI integrations: Robust Python development skills, API design, FastAPI or similar frameworks, asynchronous processing, structured output validation and automated testing.



Ability to build reliable integrations with model providers and security services.

- AWS and deployment: Hands-on AWS experience, including Bedrock, Lambda, API Gateway, S3, DynamoDB, IAM, KMS and Secrets Manager; familiarity with SageMaker, Docker and containerized workloads. Working knowledge of Git, CI/CD, infrastructure as code and monitoring.

Key responsibilities

- Define the AI security approach: Identify relevant threats, determine where controls should operate, and choose appropriate combinations of deterministic rules, machine learning and LLM-based evaluation.

- Build security capabilities: Personally contribute to attack-testing workflows, detection engines, data-protection controls, model integrations and security evidence generation.

- Own secure architecture and data flows: Define how prompts, datasets, model artifacts, responses and evidence move through the system, including where they are inspected, stored and protected.

- Validate security effectiveness: Establish representative attack and benign test datasets, benchmark detection quality, investigate bypasses and ensure that security claims are supported by evidence.

- Balance protection with performance: Improve detection effectiveness while managing response time, inference cost, scalability and unnecessary blocking of legitimate activity.

- Lead engineering delivery: Guide the team, review critical implementations, resolve technical blockers and establish testing and release standards. Communicate security risks, technical tradeoffs and delivery priorities clearly to leadership.

📌 Head of AI Security Engineering & Architecture (Vijayawada)
🏢 Logan Data
📍 Vijayawada