AI Engineer

VODA.ai is seeking a full-time AI Engineer. As part of the team, you will be responsible for building and integrating AI-powered applications and services that enhance our platform capabilities. Day-to-day tasks may include integrating APIs for external LLMs such as OpenAI, Anthropic, and Bedrock, deploying open-source models like Llama in scalable environments, contributing to our in-house agent service and its integration into our web application, extending ReAct-Loop workflows with tool calling, prompt engineering to improve system behavior, developing functional primitives for real-time web GIS interactions, designing and running evaluations against ground-truth datasets, and optimizing latency, response quality, and cost tradeoffs across AI systems. Qualifications 3+ years of experience building production software in Python, with a strong grasp of modern tooling (type hints, async, pydantic, FastAPI or similar) Hands-on experience integrating LLM APIs (OpenAI, Anthropic, Bedrock, or equivalents) into real applications; not just notebooks Practical experience with agentic patterns: ReAct loops, tool/function calling, structured outputs, and prompt engineering for reliability Solid software engineering fundamentals: API design, testing, version control, code review, and CI/CD Experience designing and running evaluations for LLM systems against ground-truth datasets, and reasoning about quality, latency, and cost trade-offs Python, FastAPI, gRPC/protobuf, and Postgres experience Expertise in AI frameworks like LangGraph, BAML, MLflow, and strong understanding of RAG and vector search patterns (Bedrock preferred) This is right for you if you: Enjoy building scalable AI-powered systems and solving complex engineering challenges Thrive in collaborative environments and enjoy experimenting to improve AI performance and user experience Are comfortable working across the stack enough to integrate LLM services into a web application and collaborate effectively with frontend engineers Have familiarity with APIs from OpenAI, Anthropic, and Bedrock Have experience with AWS or other cloud-native development environments and Docker Have experience deploying open-source models (Llama, Mistral, Qwen, etc.) in scalable environments using vLLM, TGI, SageMaker, Bedrock custom deployments, or similar Have exposure to GIS data or geospatial applications (PostGIS, GeoJSON, mapping libraries) Who we are VODA.ai uses Artificial Intelligence (AI) to help water utilities predict which of their pipes are going to fail or have lead. There are 240,000 pipe breaks each year and millions of people without fresh water, so we are doing something about it. Our software uses our proprietary, built-in-house AI and is used by water utilities and municipalities to help them create better engineering and operations decisions. We’re built by water people for water people. VODA.ai was founded in 2017 and is headquartered in Boston, MA. We have a multi-award-winning solution and customers worldwide. Read more about us at voda.ai/about.