Candent is hiring a Senior AI Security Developer with 3 to 5 Yrs of experience to secure AI-powered systems across multi-cloud LLM platforms. You will identify and remediate AI-specific vulnerabilities using automation, implement Automated assessment and security guardrails across Azure OpenAI, AWS Bedrock, Anthropic Claude, GCP Vertex AI, and OpenAI integrations, and ensure all AI workloads aligned automatically with OWASP Top 10 for LLMs, NIST AI RMF, and emerging regulatory frameworks by the application.

- Identify, assess, and remediate AI-specific vulnerabilities — prompt injection, jailbreaks, insecure output handling, training data poisoning, and model inversion

- Perform automated AI asset discovery across enterprise environments — build Python and GoLang scanners to enumerate LLM endpoints, AI services, model deployments, API gateways, and embedded AI components across Azure, AWS, and GCP

- Automate AI topology mapping — develop Python/GoLang agents to continuously discover and document model dependencies, data flows, integration points, trust boundaries, and exposure surface across multi-cloud AI estates

- Classify discovered AI assets by risk tier — automated pipelines tag and score public-facing models, internal agents, RAG pipelines, fine-tuned models, and third-party AI integrations

- Build automated vulnerability detection and remediation tooling in Python and GoLang — scanners, fuzzing harnesses, guardrail middleware, and self-healing fix pipelines for identified AI security issues

- Implement and maintain security guardrails across multi-cloud LLM platforms — Azure OpenAI, AWS Bedrock, Anthropic Claude, GCP Vertex AI, and OpenAI

- Design and enforce OWASP Top 10 for LLMs controls across all AI application layers

- Build secure RAG pipelines — enforce retrieval-time access control,



tenant boundary isolation, and PII redaction before context injection

- Conduct red-teaming and adversarial testing of LLM-integrated applications — automated and manual

- Align AI system design and deployment with NIST AI RMF, NIST SP 800-218A, EU AI Act, and Microsoft RAI standards

- Integrate all automated discovery, vulnerability detection, and fix pipelines into CI/CD — Python and GoLang based, with automated prompt injection testing, output validation, and policy enforcement gates

- Monitor LLM interactions via observability tools (Langfuse, Arize Phoenix) for anomalous behaviour and data leakage

- Collaborate with architecture and compliance teams to produce AI risk assessments and mitigation reports for enterprise clients

- Strong expertise in Python (5+ years) for automation, AI integrations, security tooling, and backend development.

- Hands-on experience in GoLang for building scalable security tools, microservices, and AI security platforms.

- Extensive experience with Azure OpenAI SDK, including GPT-4, Embeddings, DALL·E, Assistants, and enterprise AI integrations.

- Proven experience with AWS Bedrock SDK, leveraging Titan, Claude, Llama, and other foundation models.

- Solid knowledge of Anthropic Claude SDK, prompt engineering, tool usage, API integrations, and model governance.

- Experience with Google Vertex AI SDK, including Gemini, PaLM, model deployment, and AI application development.

- Expertise in OpenAI SDK, ChatGPT APIs, Function Calling, Assistants, Embeddings, and RAG-based applications.

- Hands-on implementation of Retrieval-Augmented Generation (RAG) architectures, including chunking, vectorization, retrieval optimization, re-ranking, and vector databases.

- Strong understanding of Prompt Engineering, AI model evaluation, and secure LLM application development.

📌 AI Security Developer (Pune)
🏢 Candent
📍 Pune