Responsibilities

- Define secure application architectures covering web, mobile, APIs, AI services, data stores, models, and enterprise integrations, focusing on Security Flaws, Vulnerabilities & Risks.
- Identify and mitigate traditional application vulnerabilities as well as AI-specific risks such as prompt injection, data leakage, RAG poisoning, model misuse, and excessive agent privileges.
- Design secure RAG pipelines with proper document validation, access control, retrieval security, vector-store protection, and response validation.
- Evaluate and select LLMs, embedding models, rerankers, vector databases, and graph technologies based on security, quality, scale, latency, and cost.
- Compare and design solutions using platforms such as Neo4j, Pinecone, Weaviate, Milvus, Azure AI Search, Chroma, pgvector, or similar technologies.
- Define production deployment, MLOps/LLMOps, model monitoring, evaluation, drift detection, retraining, versioning, and rollback strategies.
- Establish Responsible AI, model governance, explainability, auditability, compliance, and human oversight.
- Stay aware of emerging AI security threats and industry incidents and translate those learnings into practical architecture and security controls.

Key Skills

- Application, API, mobile, and Security.
- AI Security, Threat modelling,



prompt-injection protection, and secure AI design.
- LLMs, RAG, embeddings, semantic search, hybrid search, GraphRAG, and AI agents.
- Vector and graph databases, including selection and architecture trade-offs.
- Data architecture, data governance, privacy, PII protection, lineage, and access control.
- MLOps/LLMOps, model deployment, model registry, monitoring, drift detection, and evaluation.
- Model selection, benchmarking, hallucination/groundedness evaluation, and production quality controls.
- IAM, RBAC/ABAC, encryption, secrets management, network isolation, and secure SDLC.
- Responsible AI, model governance, bias, explainability, compliance, and auditability.
- Kubernetes/OpenShift, containers, GPUs, model serving, and enterprise observability.
- Robust communication, technical leadership, and architecture decision-making.

Experience

- 10 years of software, architecture, data, Security or AI engineering experience.
- Experience designing enterprise-grade AI solutions, preferably in regulated, government, banking, telecom, or other high-security environments.
- Experience taking AI solutions from data ingestion and model selection through deployment, monitoring, governance & production operations in a secure manner.

📌 AI Security Engineer (Bengaluru)
🏢 Arohana Tech Solutions Private
📍 Bengaluru