Agentic Systems Engineer

Kantiv (Joist AI) is hiring an Agentic Systems Engineer to build next-generation agentic applications that streamline proposal writing for the Architecture, Engineering, and Construction (AEC) industry. This is a full-time, remote opportunity in India for engineers with 2–4 years of production software development experience. Responsibilities include building modular AI agents, implementing short-term and long-term memory, integrating tools, MCP servers, and agent skills, and developing retrieval and evaluation systems. Candidates should have strong Python programming skills, clean software architecture practices, and a solid understanding of RAG, prompting patterns, tool use, structured outputs, streaming, and context management. Experience with modern agent frameworks, LLM evaluation, testing, observability, and production trace analysis is important. Hands-on experience with Langfuse, LangSmith, or equivalent LLM observability tools is expected. Experience with embeddings, vector databases, hybrid retrieval, rerankers, end-to-end LLM evaluations, and LangGraph is particularly valuable. The selection process typically takes two weeks and includes an introductory Zoom call, a Python and agentic coding assessment, a project deep dive and presentation, a Generative AI and LLM fundamentals interview, and a culture-fit interview.