over 1 year ago
Base Salary
$120k - $250k/yr
Responsibilities
- Optimize ASR, LLMs, and TTS for accurate performance in diverse and noisy environments.
- Fine-tune LLMs using RAG, reinforcement learning, and prompt engineering for context-aware conversations.
- Integrate AI components into autonomous agents for scheduling, order-taking, and issue resolution.
- Build human-in-the-loop and automated monitoring systems to detect anomalies and improve models using real-world feedback.
- Develop pipelines that construct knowledge graphs from business data.
- Scale models across GPU/TPU clusters and edge devices while minimizing latency.
- Manage experimentation, training, and optimized production inference.
- Lead evaluations, error analysis, and iterative improvements for robustness and scalability.
- Collaborate with product and customer teams to balance research innovation with practical usability.
- Publish research, contribute to open source, and present at industry conferences.
Requirements
- 3–7+ years deploying impactful machine-learning models, ideally in voice, NLP, knowledge graphs, or agent systems.
- Deep knowledge of speech recognition, language models, reinforcement-learning or dialogue systems, text-to-speech, ontology systems, or agent orchestration.
- Proficiency in PyTorch or JAX; CUDA and Triton optimization experience is preferred.
- Experience minimizing latency and resource use on GPUs, TPUs, or edge hardware.
- A strong data-driven approach with measurable improvements.
- BS, MS, or PhD in Computer Science, Electrical Engineering, Mathematics, or equivalent practical expertise.
- Passion for creating intuitive, helpful, and frustration-free AI experiences.
Benefits
- Work in person five days per week at the NYC office.
- Competitive salary and meaningful equity.
- Full health, dental, vision, 401(k), and life insurance.
- Unlimited PTO.
- Tools budget, coffee budget, and a flexible whatever-you-need-to-be-great budget.
- Small, high-trust team with a founding-team feel and fast feedback loops.
Tech Stack
Categories
AI ResearchML Engineering
About Maple
Maple builds AI voice agents that answer calls, take orders, book appointments, and handle payments for restaurants and other local businesses, syncing with existing POS and booking systems. It sells this as a software service that powers phone ordering, reservations, drive-thru, and kiosk experiences. Founded in 2023 and headquartered in New York, the company is privately held.