Voice AI Engineer (Real-Time Speech)

<p>We are looking for a&nbsp;<strong>Voice AI Engineer (Real-time speech)</strong> to join our team!&nbsp;</p> <p><strong>Our client</strong> is an Azerbaijani telecommunications company and Azerbaijan's largest mobile network operator. The main products are: Fixed telephony, Mobile telephony, Internet services, Wireless broadband, and Value-added services. The primary goal is to accelerate the client’s Data &amp; AI initiatives via a secure, hybrid cloud foundation on AWS while systematically modernizing the IT estate as part of the AWS MAP 2.0 program.</p> <p><strong>Key Project Objectives include:</strong></p> <ul> <li><strong>Cloud Foundation &amp; Landing Zone: </strong>Deploy target hybrid network architectures, establishing a secure AWS Landing Zone Accelerator (LZA) and hybrid Data/AI platforms on AWS.</li> <li><strong>Security, Compliance &amp; Sovereignty: </strong>Operationalize on-premises data de-identification and Format Preserving Encryption (FPE) tokenization (achieving zero raw PII in the cloud), fully adhering to Azerbaijani Personal Data Law No. 998-IIIQ and Critical Information Infrastructure Rules (Resolution No. 229).</li> <li><strong>AI Chatbot &amp; Real-Time Voicebot Implementation: </strong>Develop and operationalize a flagship Customer Care Voicebot (STT → LLM → TTS pipeline) and Agentic Chatbot targeting &lt; 2.0s conversational voice latency and ~200 rps throughput as the first hybrid-setup consumer.</li> </ul> <p>&nbsp;</p> <p><strong>Responsibilities:</strong></p> <ul> <li>Design, build, and operationalize end-to-end real-time <strong>STT → LLM → TTS (Speech-to-Text / LLM / Text-to-Speech)</strong> voicebot pipelines on AWS, optimizing for streaming speech-to-text, first-token LLM generation, and first-audio TTS synthesis.</li> <li>Deploy and maintain production customer-trained <strong>Whisper (Azerbaijani ASR)</strong> and <strong>Azerbaijani TTS</strong> models as low-latency real-time endpoints on <strong>Amazon SageMaker</strong> and specialized GPU node pools (NVIDIA A100/L40S).</li> <li>Implement and manage the&nbsp;<strong>Bedrock Proxy Gateway</strong> on EKS for multi-model routing, priority queuing via Redis Sorted Sets, cost caps, and high-availability serving targeting ~200 rps without API throttling.</li> <li>Integrate voicebot and chatbot decision engines with core enterprise telephony and CVM platforms, including&nbsp;<strong>Avaya</strong> (voice telephony), <strong>Genesys</strong> (digital chat/omnichannel), and <strong>Pelatro</strong> (CVM offer decisioning and uplift models).</li> <li>Establish&nbsp;<strong>LLMOps &amp; MLOps</strong> pipelines using <strong>Amazon SageMaker Pipelines</strong> and <strong>MLflow</strong> for experiment tracking, model versioning, prompt/agent registries, automated evaluation harnesses, and RAG knowledge base retrieval.</li> <li>Build call and chat transcription pipelines to ingest, transcribe, and extract real-time insights (churn risk, dissatisfaction, intent, lead signals) into downstream decision layers.</li> <li>Enforce data sovereignty and privacy controls by integrating on-premises&nbsp;<strong>Format Preserving Encryption (FPE)</strong> and tokenization wrappers into ML pipelines so zero raw PII enters AWS cloud environments.</li> <li>Define NFR baselines, dialogue flows, voicebot persona, turn-taking, and fallback/escalation logic to guarantee conversational round-trip latency &lt; 2.0 seconds.</li> <li>Automate ML deployment workflows using&nbsp;<strong>GitLab CI/CD</strong> and Infrastructure-as-Code (<strong>Terraform</strong> or <strong>AWS CDK</strong>), establishing observability and FinOps spend/anomaly monitoring via <strong>Amazon CloudWatch</strong> and <strong>Splunk</strong>.</li> </ul> <p>&nbsp;</p> <p><strong>Requirements:</strong></p> <ul> <li>4+ years of hands-on experience with machine learning and Speech Processing with a primary focus on real-time conversational AI, <strong>ASR (STT)</strong>, and <strong>TTS</strong> voice pipelines.</li> <li>Deep expertise with <strong>Amazon SageMaker</strong> (real-time GPU inference endpoints, Pipelines, Feature Store, Model Registry) and <strong>Amazon Bedrock</strong> (AgentCore, Bedrock Guardrails, Knowledge Bases).</li> <li>Proven track record in streaming speech inference, speech synthesis, and low-latency audio processing.</li> <li>Strong experience in <strong>GPU optimization</strong> and containerized orchestration (NVIDIA A100/L40S, AWS EC2 GPU instances, Docker, Kubernetes/EKS).</li> <li>Solid understanding of contact center and telephony platform integrations (<strong>Avaya</strong>, <strong>Genesys</strong>) and real-time decisioning interfaces.</li> <li>Proficient in Python, Redis (priority queuing &amp; caching), and data security/privacy (FPE tokenization, handling sensitive/PII data).</li> </ul> <p>&nbsp;</p> <p><strong>Nice-to-have skills:</strong></p> <ul> <li><strong>AWS Certified Machine Learning – Specialty</strong> or <strong>AWS Certified Solutions Architect</strong>.</li> <li>Hands-on experience with EMR-on-EKS, Apache Iceberg, or MSK (Kafka) streaming pipelines.</li> </ul> <p>&nbsp;</p> <p><strong>Soft Skills &amp; Team Fit:</strong></p> <ul> <li>Strong critical thinking, problem-solving, and analytical skills with ownership of mission-critical, low-latency deliverables.</li> <li>Excellent communication and collaboration skills to work closely with cross-functional teams (AI Architects, Data Engineers, CC SMEs, and Security/Compliance).</li> <li>Results-oriented, proactive mindset with strong ownership within an Agile / Scrum framework.</li> <li>Upper-Intermediate+ English level (written and spoken).</li> </ul> <p><br><br></p><div class="content-conclusion"><p><strong>We offer*:</strong></p> <ul> <li>Flexible working format - remote, office-based or flexible</li> <li>A competitive salary and good compensation package</li> <li>Personalized career growth</li> <li>Professional development tools (mentorship program, tech talks and trainings, centers of excellence, and more)</li> <li>Active tech communities with regular knowledge sharing</li> <li>Education reimbursement</li> <li>Memorable anniversary presents</li> <li>Corporate events and team buildings</li> <li>Other location-specific benefits</li> </ul> <p>*not applicable for freelancers</p></div>