Description

Lead the design and development of an AI voice agent platform, owning the architecture, speech processing integration, and model performance optimization.

This role is remote.

Responsibilities

• Design and build the architecture for an AI voice agent platform.
• Customize open-source LLMs using Hugging Face to meet specific business requirements.
• Integrate TTS and STT functionalities for audio and speech processing.
• Implement embeddings and vector databases to optimize data storage and retrieval.
• Use call recordings and transcripts to re-train models for improved intelligence.

Required Skills

• Hands-on experience in AI and Machine Learning.
• Expertise in audio and speech processing, including TTS and STT.
• Experience with LLMs and Hugging Face for model training and customization.
• Proficiency with Eleven Labs, Play.ht, and Deepgram.
• Practical knowledge of embeddings and vector databases.
• Experience with PyTorch, RAG, and AI/ML workflows.
• Ability to perform hands-on coding and architectural design.

Preferred Skills

• Familiarity with GPTCache or similar technologies for caching answers and audio files to reduce latency.
• Knowledge of GPU-based hosting companies.

Key Skills

Education

Not specified