AI / LLM Researcher

We are a well-funded, early-stage startup, seeking a talented and motivated AI/LLM Researcher to join our team. The focus of this role is to develop advanced AI agents aimed at automating and optimizing low-skill human tasks, with a special emphasis on document processing, data extraction, text and email comprehension, and related workflows. You will be responsible for researching, designing, and deploying AI solutions that can replace repetitive manual tasks, improving efficiency and scalability for organizations.

Key Responsibilities:

AI Agent Development:

• Design, develop, and implement AI agents using best of breed LLMs to automate document processing tasks across enterprise workflows.
  
  

• Build machine learning pipelines to automate content-heavy processes for enterprise businesses.
  
  

• Fine-tune and optimize pre-trained LLMs to accurately interpret, classify, and process large volumes of documents, improving speed and accuracy.
  
  

• Partner with product, engineering, and business teams to identify high-value automation opportunities and integrate AI-driven solutions into our operations.
  
  

• Develop metrics and conduct extensive testing to ensure reliability and efficiency of the AI systems in real-world scenarios.
  
  

• 2+ years of experience with large language models and NLP techniques.
• Background in developing AI agents for content heavy tasks.
• Experience with multiple LLM Vendors, RAG or Fine Tuning OSS or Commercial models..
  
  

• Proficiency in Python and machine learning libraries such as TensorFlow or PyTorch.
• Deep understanding of LLM architectures and NLP applications like entity recognition, document classification, and summarization.
• Experience with model grounding techniques, retrieval-augmented generation (RAG), and deploying robust AI models in production environments.
• Skills in designing and deploying AI models through APIs, containerization (Docker, Kubernetes), and cloud platforms (AWS, GCP, Azure).
• Understanding of best practices for continuous monitoring, evaluation, and iterative improvement of deployed models.