Applied AI Engineer

Role Overview: As an Applied AI Engineer at Instant Markets, a venture of Instant Systems Inc., you will be responsible for leveraging the latest in Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and Natural Language Processing (NLP) to develop cutting-edge applications such as intelligent assistants, semantic search, and AI copilots. Your role will involve designing and implementing LLM-based applications, building RAG pipelines, applying NLP techniques, fine-tuning pretrained LLMs, developing ML pipelines, working with APIs, collaborating with cross-functional teams, and monitoring model performance. Key Responsibilities: - Design and implement LLM-based applications using models like GPT, BERT, T5, LLaMA, etc. - Build RAG pipelines that combine retrieval systems with generative models for accurate outputs. - Apply NLP techniques to solve tasks like question answering, summarization, and classification. - Fine-tune pretrained LLMs and customize them for specific domains or use cases. - Develop and maintain end-to-end ML pipelines, from data collection to model deployment and monitoring. - Work with APIs like OpenAI's ChatGPT, Hugging Face Transformers, and LangChain to build user-facing GenAI solutions. - Collaborate with cross-functional teams to translate AI research into scalable product features. - Leverage cloud platforms and tools like MLflow, Airflow, Docker, and Kubernetes. - Monitor model performance, detect drift, and apply retraining or tuning strategies when needed. Qualifications Required: - 4 years of experience in Machine Learning, Deep Learning, and NLP. - Strong hands-on experience with LLMs like GPT, BERT, LLaMA, including fine-tuning and prompt engineering. - Practical knowledge of Retrieval-Augmented Generation (RAG) architecture. - Experience integrating and building with ChatGPT, OpenAI APIs, or similar LLM services. - Proficiency in Python and ML libraries like PyTorch, TensorFlow, Transformers (Hugging Face). - Solid understanding of semantic search, vector embeddings, and vector databases like FAISS, Pinecone, Weaviate. - Familiarity with LangChain, Haystack, or similar frameworks. - Experience with deploying models using Docker, Kubernetes, MLflow, and cloud platforms like AWS, GCP, or Azure.,