AI Engineer - RAG / LLM

We are seeking an AI Engineer with 3 - 8 years of experience to design, build, and deploy end-to-end AI solutions with a focus on Generative AI (GenAI), Retrieval-Augmented Generation (RAG) pipelines, copilots, and applied AI systems that deliver measurable business impact. The role involves architecting scalable AI platforms, fine-tuning large language models (LLMs) using PEFT techniques (LoRA, QLoRA, Unsloth), and building LLM Ops pipelines for model lifecycle management. The ideal candidate combines AI engineering, MLOps, and cloud expertise, with experience across NLP, computer vision, and multimodal AI, and a strong ability to deliver enterprise-grade AI solutions for banking.

The core responsibilities for the job include the following :

Primary :

• Build and deploy AI-powered applications(chatbots, copilots, automation flows) for banking operations and customer service.
• Design and implement RAG pipelines and AI agents for secure financial data insights.
• Fine-tune and optimize LLMs using LoRA, QLoRA, and other PEFT techniques.
• Develop end-to-end LLMOps pipelines(training, evaluation, deployment, monitoring).
• Expose backend APIs / microservices to integrate LLMs into banking platforms.
• Deploy scalable AI models on AWS / Azure with Docker, Kubernetes, CI / CD, while ensuring security and :
• Collaborate with product managers, data scientists, and compliance teams to translate business needs into AI solutions.
• Create reusable AI components, SDKs, and templates for faster adoption.
• Support data engineering pipelines(ETL / ELT, Spark, Airflow).
• Conduct A / B testing and feedback collection for continuous model improvement.
• Explore emerging GenAI tools, open-source models, and fine-tuning :
• Mentor and lead a team of AI engineers, promoting best practices in LLMOps and GenAI.
• Drive cross-functional collaboration to align AI solutions with business goals.
• Oversee project delivery, resource planning, and ensure quality standards.

Key Success Metrics :

Requirements :

(ref : hirist.tech)