AI Engineer
Deloitte (LATAM)
REMOTE
Experience - 7+ years
Role Overview
As an AI Voice & LLM Integration Developer, you will play a pivotal role in building and optimizing
Your work will focus on:
* Integrating GPT's ASR (speech-to-text) and TTS (text-to-speech) APIs into our platform.
* Handling real-time voice streaming from VoIP/SIP systems.
* Optimizing low-latency communication between AI models and telephony platforms (e.g., FreeSWITCH, WebRTC).
* Supporting PBX call routing & handoWs with AI agents.
* Ensuring high-quality, natural-sounding AI-driven voice interactions.
Key Responsibilities
* Design and develop real-time voice processing pipelines for AI-driven conversations.
* Integrate OpenAI's Whisper (ASR) and TTS APIs with VoIP and WebRTC systems.
* Develop low-latency audio streaming middleware between telephony providers(Telnyx, Bandwidth) and GPT.
* Work with SIP and WebRTC to ensure seamless audio transmission.
* Implement PBX call routing and AI handoW mechanisms (e.g., transferring AI-handled calls to human agents).
* Optimize AI voice latency and response times for a real-time, natural experience.
* Collaborate with VoIP engineers to integrate AI capabilities into SIP-based systems.
* Design APIs to manage AI-driven IVR workflows and customer interactions.
* Ensure compliance with telephony regulations (e.g., STIR/SHAKEN, call recording laws).
* Monitor system performance and implement scalability strategies for voice interactions.
Required Qualifications
* AI & LLM Integration
o Experience integrating GPT APIs (ChatGPT, Whisper, TTS engines) for voice applications.
o Understanding of real-time AI voice processing and conversational AI workflows.
* Voice & Audio Streaming
o Experience working with low-latency, real-time audio streaming.
o Familiarity with WebRTC, SIP/RTP, and VoIP streaming.
* Programming & Middleware Development
o Proficiency in Python, Node.js, or Go for API and middleware development.
o Experience developing RESTful APIs and real-time streaming solutions.
* VoIP & Telephony Knowledge
o Familiarity with SIP, FreeSWITCH, Asterisk, or other PBX platforms.
o Experience working with SIP trunking providers (Telnyx, Bandwidth, Flowroute, etc.).
* Performance & Optimization
o Ability to reduce latency in AI-driven conversations.
o Experience with audio codec optimization (G.711, Opus, etc.).
* One or more of the following additional qualifications:
o Experience with STT/ASR models beyond GPT (Google Speech-to- Text, Deepgram, Kaldi).
o Background in signal processing or audio engineering.
o Familiarity with cloud-based voice solutions (AWS Connect, Twilio Voice, Dialogflow CX).
o Knowledge of containerization (Docker, Kubernetes) for scaling AI voice workloads.
o Experience with multi-tenant architectures for voice AI.