Role: Sr Agentic AI Engineer Location: Atlanta, GA (Hybrid – primarily onsite) ONLY LOCALS Experience: 10+ Years We are seeking a Senior AI Engineer with strong expertise in Python, TypeScript, and Java, combined with hands-on experience working with agentic AI systems like Google AgentSpace, LangGraph / LangChain, CrewAI, and AWS Bedrock. The ideal candidate is proficient with state-of-the-art LLMs including Claude, GPT-4o, Haiku, LLaMA, Nova, and has experience implementing real-time speech-to-speech AI models, including OpenAI’s Realtime API and Amazon's Novasonic. You will lead the design and deployment of intelligent, autonomous agents for real-world applications—leveraging cutting-edge tools to build dynamic, multimodal, and scalable AI solutions. Key Responsibilities: Design, develop, and deploy agentic AI systems using frameworks like LangGraph, LangChain, CrewAI, and Google AgentSpace. Build and integrate applications on AWS Bedrock using foundational models from multiple providers. Implement and optimize workflows using LLMs (Claude, GPT-4o, LLaMA, Haiku, Nova, etc.) for reasoning, summarization, conversation, and decision-making tasks. Integrate real-time speech-to-speech AI models (e.g., OpenAI Realtime API, Novasonic) for live conversational AI experiences. Develop backend services and APIs using Python, TypeScript, and Java for scalable AI applications. Collaborate cross-functionally with product, data, and research teams to define and deliver AI-first products. Evaluate and fine-tune foundation models for specific tasks and optimize performance across modalities (text, voice, audio). Stay updated on emerging trends in AGI/agentic systems, speech models, and foundation model ecosystems. Required Skills & Qualifications: 10+ years of experience in software development with 5+ years of strong programming expertise in Python, TypeScript, and Java. Deep hands-on experience with LangGraph, LangChain, CrewAI, Google AgentSpace, or similar frameworks. Experience with AWS Bedrock and working with multimodal and multi-vendor foundation models. Familiarity with state-of-the-art LLMs: Claude (Anthropic), GPT-4o (OpenAI), LLaMA (Meta), Haiku (Mistral), Nova (Amazon). Experience integrating real-time speech-to-speech systems including OpenAI’s Realtime API, Novasonic, or similar. Strong understanding of prompt engineering, context management, vector stores, and memory-based agent behavior. Familiarity with cloud infrastructure, APIs, containerization (Docker), and deployment pipelines. Excellent problem-solving, system design, and communication skills. Nice to Have: Experience working with multi-agent systems for planning, retrieval-augmented generation (RAG), and tool use. Experience with streaming architectures or edge deployment for real-time applications. Contributions to open-source AI projects or publications in applied AI research.