Applied AI Researcher

As a Senior AI Engineer - Voice AI / Autonomous Agents at Spyne, you will have the opportunity to own and build Spynes in-house voice bot stack. In this high-impact individual contributor role, you will be at the intersection of LLMs, ASR/TTS, and voice UX, focusing on creating deeply human, latency-optimized conversations between auto dealerships and their customers. Your main responsibilities will include: Voice AI Stack Ownership: Building and owning the end-to-end voice bot pipeline, including ASR, NLU, dialog state management, tool calling, and TTS to deliver a natural, human-like conversation experience. LLM Orchestration & Tooling: Architecting systems using MCP (Model Context Protocol) to mediate structured context between real-time ASR, memory, APIs, and the LLM. RAG Integration: Implementing retrieval-augmented generation to ground responses using dealership knowledge bases, inventory data, recall lookups, and FAQs. Vector Store & Memory: Designing scalable vector-based search for dynamic FAQ handling, call recall, and user-specific memory embedding. Latency Optimization: Engineering low-latency, streaming ASR + TTS pipelines and fine-tuning turn-taking models for natural conversation. Model Tuning & Hallucination Control: Using fine-tuning, LoRA, or instruction tuning to customize tone, reduce hallucinations, and align responses to business goals. Instrumentation & QA Looping: Building robust observability, running real-time call QA pipelines, and analyzing interruptions, hallucinations, and fallbacks. Cross-functional Collaboration: Working closely with product, infra, and leadership to scale this bot to thousands of US dealerships. To be successful in this role, you should possess: Architect-level thinking: Understanding how ASR, LLMs, memory, and tools fit together and having the ability to design modular, observable, and resilient systems. LLM Tooling Mastery: Implementing tool calling, retrieval pipelines, function calls, or prompt chaining across multiple workflows. Fluency in Vector Search & RAG: Knowing how to chunk, embed, index, and retrieve, while avoiding prompt bloat and token overflow. Latency-First Mindset: Debugging token delays, understanding the cost of each API hop, and optimizing round-trip time to maintain human-like interactions. Grounding > Hallucination: Tracing hallucinations back to weak prompts, missing guardrails, or lack of tool access and effectively addressing them. Prototyper at heart: Being unafraid of building from scratch and iterating quickly, utilizing open-source or hosted tools as necessary. The ideal candidate will have: 5+ years of experience in AI/ML or voice/NLP systems with real-time experience. Deep knowledge of LLM orchestration, RAG, vector search, and prompt engineering. Experience with MCP-style architectures or structured context pipelines between LLMs and APIs/tools. Experience integrating ASR (Whisper/Deepgram), TTS (ElevenLabs/Coqui), and OpenAI/GPT-style models. Solid understanding of latency optimization, streaming inference, and real-time audio pipelines. Hands-on experience with Python, FastAPI, vector DBs (Pinecone, Weaviate, FAISS), and cloud infra (AWS/GCP). Strong debugging, logging, and QA instincts for hallucination, grounding, and UX behavior. Working at Spyne offers real-world AI impact at scale, a high-performing team that balances speed with technical depth, high autonomy and visibility from day one, rapid career acceleration, access to MacBook and all necessary tools and compute, a flat structure with real work focus, and no BS. Join us in redefining how cars are marketed and sold with cutting-edge Generative AI.,