Job
Description
Keytag.ai are looking for AI/LLM engineers with to join our team building a new global LLM operations and agent framework for a large pharmaceutical organization. We are rolling out an eco-system of LLM Agents to generate documents, to automate workflows, and to perform new tasks across the organization. We are using cloud hosted and on-prem models as appropriate to the business area, and combining RAG, fine-tuning and some training. This a great opportunity to work with all aspects of Generative AI, in a startup mode with the security of company that has fixed customer contracts for the coming years. Core Skills Required Experienced in developing RAG based LLM solutions, handling all aspects from document splitting and chunking strategies, domain specific retrieval strategies, prompt development, performance tuning and production troubleshooting. Good understanding of LangChain and LangGraph libraries. Strong Python development and code design skills. Experienced in all aspects of data engineering and operations Ability to develop in Linux Environment, VSCode, github. Self starter keen to learn, enjoys delivering solutions, happy to give opinions and influence the design as well as the development of solutions. Additional Desirable Skills : Development of domain specific and/or task specific Agents in langgraph or another library. Experience designing and developing multi-agent flows. Experience with Langfuse, DeepEval or DPSy Fine-Tuning of models (and sample generation) to understand an organizations vernacular and services. Experience with multiple model providers eg Openai, Azure Openai, Lllama, Falcon. Experience using Neo4j, Redis, Fastapi, Airflow. What the role will involve : Working with the business on specific use cases to develop Agents to perform specific tasks and fit within our existing agent framework. Use cases are a variety of question answering and summarization, document and asset generation, workflow automation, and brand new tasks to replace complex human analysis. Enhancing our agent framework and ecosystem as requirements change and Gen AI technology advances. Creating a suite of models that can be used by the business; fine-tuned organization LLM and smaller task/domain specific llms. Using both cloud and on-prem open source models as appropriate. Enhancing our MLOps practices and improving our testing automations Sourcing and mapping data from a wide range of services including neo4j and relational databases, SAAS services, security and general business document stores. It is a remote role but there is plenty of interaction with the rest of the team, with daily standups and shared tasks, as well as directly working with customers on projects to discuss changes, implement, troubleshoot solutions etc. Benefits Remote work opportunity, providing flexibility and work-life balance. Medical insurance coverage for you and your family. Join a friendly and well-organized team with an open communication style. Continuous opportunities to grow your technical and soft skills.
