Senior ML Engineer - Text-to-Speech & Scalable Inference

ConnexAI; Manchester; On-site; GBP 90,000 - 140,000; Full time

Job summary

ConnexAI in Manchester, United Kingdom, is seeking a Senior Machine Learning Engineer to deploy state‑of‑the‑art Text‑to‑Speech models and scale TTS systems for production. You will optimise pipelines for GPU performance and ensure reliable inference at scale.

You will work with PyTorch and Hugging Face transformers, integrate LLM inference servers and complex production pipelines, and push the boundaries of neural audio codecs like Encodec.

Qualifications

• MSc or PhD in Computer Science or a related field.
• 3–5 years of hands‑on experience deploying and scaling ML solutions in production.
• Proven experience deploying and optimising LLMs/transformers in production.
• Knowledge of LLM inference servers (e.g., Triton, TensorRT, TorchServe).
• Experience with GPU scaling for large‑scale ML models.
• Expertise in deploying complex ML pipelines in production environments.
• Proficiency with PyTorch and Hugging Face transformers.
• Experience with neural audio codecs (e.g., Encodec).
• Background in Text‑to‑Speech (TTS) development.
• Experience with RVQ, GANs, and diffusion models.

Responsibilities

• Collaborate closely with the TTS team to deploy and scale advanced models in production environments.
• Lead efforts in optimising TTS pipelines for performance and scalability, particularly focusing on GPU utilisation.
• Implement and maintain LLM and transformer models, ensuring efficient inference at scale.
• Integrate and manage LLM‑based inference servers like Triton, TensorRT, or TorchServe to streamline deployment.
• Work on deploying complex pipelines in production, ensuring seamless integration with existing systems.

Skills

LLM deployment

PyTorch

Hugging Face

GPU scaling

Transformer models

TTS development

Inference optimization

TorchServe/Triton

VPUs/ GPUs optimization

Education

MSc or PhD in Computer Science or related field

Tools

Triton

TensorRT

TorchServe

Job description

ConnexAI in Manchester, United Kingdom, is seeking a Senior Machine Learning Engineer to deploy state‑of‑the‑art Text‑to‑Speech models and scale TTS systems for production. You will optimise pipelines for GPU performance and ensure reliable inference at scale.

You will work with PyTorch and Hugging Face transformers, integrate LLM inference servers and complex production pipelines, and push the boundaries of neural audio codecs like Encodec.