Key Responsibilities

• Design and optimize prompts using zero-shot, few-shot, chain-of-thought, ReAct, and self-consistency techniques
• Build structured prompt templates and refine them via prompt chaining, decomposition, and verification strategies
• Implement safety, guardrails, and compliance measures for LLM-based systems
• Build conversational flows and chatbots on top of modern LLMs (OpenAI, Anthropic, LLaMA, Mistral, Gemini)
• Develop Retrieval-Augmented Generation (RAG) pipelines using embedding models, vector databases, and chunking strategies
• Build, deploy, and optimize end-to-end ML systems including data preprocessing, feature engineering, and model development
• Integrate LLM prompts into production applications via REST APIs
• Support computer vision related projects as needed

Requirements

• 10+ years of experience implementing AI and ML use cases, primarily generative AI
• Hands-on experience with modern LLMs (OpenAI, Anthropic, LLaMA, Mistral, Gemini)
• Strong understanding of tokenization, model behaviors, reasoning patterns, and evaluation frameworks
• Experience with vector databases such as FAISS, Milvus, or Pinecone
• Strong Python proficiency and experience with PyTorch or TensorFlow
• Experience with cloud services and MLOps practices
• Good understanding of REST APIs, JSON, and YAML
• Familiarity with Git, version control, and experiment tracking
• Excellent communication skills
• Must be authorized to work on H1B or H4-EAD

Skills Required