Generative AI Engineer

Job Profile Specification: Senior Generative AI / Agentic AI Engineer (5–6 years)sRole summary:- Senior-level engineer (5–6 years of professional experience) focused on designing, building, and deploying production-grade generative AI and agentic-AI solutions. - Responsible for delivering secure, scalable, and business-oriented AI systems that operate on structured and unstructured data and enable AI-driven decision-makingRequired experience- 5–6 years of industry experience in software engineering and AI-related roles. - Minimum 2-3 years of direct experience with Generative AI and Large Language Models (LLMs).Key Responsibilities:- Architect, develop, test, and deploy generative-AI solutions (online/offline LLMs, SLMs, TLMs) for domain-specific use cases. - Design and implement agentic AI workflows and orchestration using frameworks such as LangGraph, Crew AI, or equivalent. - Integrate enterprise knowledge bases and external data sources via vector databases and Retrieval-Augmented Generation (RAG). - Build and productionize ingestion, preprocessing, indexing, and retrieval pipelines for structured and unstructured data (text, tables, documents, images). - Implement fine-tuning, prompt engineering, evaluation metrics, A/B testing, and iterative model improvement cycles. - Conduct/model red-teaming and vulnerability assessments of LLMs and chat systems using tools like Garak (Generative AI Red-teaming & Assessment Kit). - Collaborate with MLOps/platform teams to containerize, monitor, version, and scale models (CI/CD, model registry, observability). - Ensure model safety, bias mitigation, access controls, and data privacy compliance in deployed solutions. - Translate business requirements into technical designs with clear performance, cost, and safety constraints.Required Skills and Experience:- Strong proficiency in Python and experience with ML/AI libraries (scikit-learn, TensorFlow, PyTorch, Hugging Face ecosystem). - Hands-on experience with LLMs, RAG, vector databases, and retrieval pipelines. - Practical experience deploying agentic workflows and building multi-step, tool-enabled agents. - Experience using Garak (or similar LLM red-teaming/vulnerability scanners) to identify model weaknesses and harden deployments. - Demonstrated experience implementing content filtering / moderation systems. - Solid skills working with structured and unstructured data and advanced feature engineering. - Familiarity with cloud GenAI platforms and services (Azure AI Services preferred; AWS/GCP acceptable). - Experience building APIs/microservices; containerization (Docker), orchestration (Kubernetes). - Strong understanding of model evaluation, performance profiling, inference cost optimization, and observability. - Good knowledge of security, data governance, and privacy best practices for AI systems.Preferred / differentiating qualifications- Hands-on fine-tuning experience and parameter-efficient tuning methods. - Experience with multimodal models and retrieval-augmented multimodal pipelines. - Prior work on agentic safety, tool-use constraints, LLM application firewalls, or human-in-the-loop systems. - Familiarity with LangChain, LangGraph, Crew AI, or similar orchestration libraries.Values & behaviours- AI-first thinking: consistently seeks AI-enabled solutions to business problems. - Data-driven mindset: makes decisions based on measurable insights and metrics. - Collaboration & agility: effective contributor in cross-functional, fast-paced teams. - Problem-solving orientation: looks beyond the obvious to unlock product and business value. - Business impact focus: designs solutions with measurable outcomes and real adoption. - Continuous learning: stays current with academic research, open-source tooling, and best practices.