# AI Developer / AI Engineer **What You'll Do** As a founding AI engineer at Peppr AI, you'll take ownership of designing, developing, and deploying intelligent systems that form the core of our agentic AI platform. You'll build ML/LLM pipelines, integrate AI agents into production systems, and collaborate closely across UX, infrastructure, and backend domains to create seamless AI-powered enterprise experiences. Key responsibilities include: * Develop end-to-end AI pipelines: data ingestion, context engineering, retrieval-augmented generation (RAG), model integration, validation, and monitoring * Design and optimize low-latency voice agents, including STT, VAD, TTS, and streaming solutions using WebRTC or WebSockets * Implement production-grade AI features, integrating them with FastAPI endpoints, Weaviate vector databases, and secure authentication systems * Collaborate with frontend and backend teams to deliver AI-powered user experiences * Continuously evaluate and prototype emerging frameworks (e.g., Hugging Face, LangChain, LLaMA) to improve performance and scalability **What We're Looking For** We're seeking a builder excited to push the boundaries of AI in enterprise settings. You should have: **Required:** * 3+ years of experience in AI/ML engineering, with hands-on experience in building and deploying LLM-based systems * Strong proficiency in Python and familiarity with frameworks such as Hugging Face Transformers and LangChain * Practical experience with Weaviate or similar vector databases for RAG-based systems * Understanding of low-latency voice technologies, including streaming pipelines and real-time audio processing * Solid background in infrastructure: Docker, Kubernetes, CI/CD, monitoring, and secure API development **Preferred:** * Experience building multi-step agent orchestration or custom AI workflows * Knowledge of latency reduction techniques (e.g., model quantization, caching) * Familiarity with designing voice-first conversational UX * Strong understanding of security, data protection, and adversarial considerations in AI pipelines * A product mindset and experience shipping end-to-end features * Comfort with ambiguity and rapid iteration in a startup environment