Sr. Machine Learning Engineer

Overview In this role you will design and ship voice and text AI agents within Realm-X, AppFolio’s AI-native platform for real estate. You will help define production voice and chat agent pipelines, balancing reasoning depth with low latency across large and small models. You’ll lead a pod of ML and platform engineers, shaping agent quality through KPIs and evaluation harnesses. You’ll work closely with Product, Voice, and ML Platform teams to deliver scalable, multi-channel experiences that impact customers at scale. This is an opportunity to advance AI-driven automation in property management. Compensation / BenefitsTotal Rewards packagehybrid workbenefits for regular full-time employeesgrowth opportunitiescollaborative culturecompetitive compensation ResponsibilitiesArchitect and ship real-time voice and text agent pipelines for multi-turn interactionsBalance reasoning depth and latency across frontier LLMs and smaller modelsLead a small pod of ML and platform engineers, improving observability and incident responseDefine KPIs, evaluation harnesses, and acceptance criteria for agent qualityOptimize voice latency and cost via selective fine-tuning of small language models (SLMs)Collaborate with Product and Voice teams to ensure high-quality agent experiences Key requirementsShipped production AI agents for voice and/or text channelsExperience designing pipelines and systems, not only modelsProven fast delivery with sound engineering judgmentCollaborative, low-ego leadership and ability to uplift team membersValue work-life balance to sustain high performanceMust Have: deep experience with LangChain, LangGraph, LangSmith, LangChain Deep Agents (or equivalents)Voice stack experience with Voice-to-Voice models and traditional TTS/STT pipelines; understanding end-to-end vs modular architecturesStrong LLM fluency in reasoning, tool use, structured output, and latency trade-offsTelephony & cloud experience with Twilio and AWSExpert Python, async programming, and WebSockets for real-time streamingML fundamentals with deep learning, model evaluation, inference; Docker on AWShumilitycollaborationteam leadershipLangChain / LangGraph / LangSmith / LangChain Deep AgentsVoice-to-Voice models, TTS, STTLLM reasoning and tool use