Generative AI/LLM Engineer

Generative AI/LLM Engineer - Hyderabad

Company: Soothsayer Analytics

Location: Hyderabad, Telangana, India

Job Type: Full-Time

Number of Positions: 4

Position Overview

Soothsayer Analytics is looking for a skilled Generative AI/LLM Engineer to join our dynamic team in Hyderabad. This role focuses on building and deploying advanced AI models, with a strong emphasis on state-of-the-art technologies like Azure OpenAI GPT-4, GPT-4 Vision, and GPT-4 Omni. Proficiency in Retrieval-Augmented Generation (RAG) and experience with Vector Databases are crucial for success in this position. While fine-tuning large language models (LLMs) is considered a valuable asset, it is not a mandatory requirement. A solid understanding of deep learning model training and fine-tuning processes is essential. The ideal candidate will demonstrate an ability to rapidly acquire and implement new techniques, even without prior direct experience in all specified areas.

Key Responsibilities

• Design, develop, and deploy generative AI models leveraging GPT-4 variants (including GPT-4 Vision and GPT-4 Turbo) to meet diverse business requirements.
• Implement and optimize Retrieval-Augmented Generation (RAG) techniques for sophisticated data-driven applications.
• Develop and manage AI services using Python frameworks such as LangChain or LlamaIndex, and create APIs with FastAPI or Quart for seamless integration.
• Prioritize scalability, performance, and optimization of AI solutions within cloud environments, particularly Azure and AWS.
• Work with Vector Databases (mandatory) and optionally Graph Databases for enhanced data management strategies.
• Utilize Cosmos DB and SQL for robust data storage and management.
• Apply MLOps or LLMOps principles to automate and streamline the AI model lifecycle, encompassing CI/CD pipelines, monitoring, and ongoing maintenance.
• Implement and manage Azure Pipelines for continuous integration and deployment workflows.
• Engage in continuous research and adoption of the latest AI advancements, focusing on rapid learning and application of emerging technologies.

Required Skills and Qualifications

• Bachelor's or Master's degree in Computer Science, AI, Data Science, or a related discipline.
• A minimum of 1+ years of dedicated experience in Generative AI/LLM technologies, complemented by 5+ years of overall experience in relevant fields.
• Strong proficiency in Python and experience with frameworks like LangChain, LlamaIndex, FastAPI, or Quart.
• Expertise in Retrieval-Augmented Generation (RAG) and practical experience with Vector Databases (mandatory).
• Hands-on experience with Cosmos DB and SQL.
• Experience with LLM fine-tuning and Graph Databases are advantageous but not mandatory.
• Demonstrated experience in MLOps, LLMOps, or DevOps, with a solid grasp of CI/CD processes, automation, and pipeline management.
• Familiarity with containerization technologies like Docker or Kubernetes is beneficial.
• Experience with cloud platforms, particularly Azure or AWS, and familiarity with cloud-native AI services.
• Excellent problem-solving capabilities and a proactive attitude towards quickly learning new AI trends and best practices.