Build the future of
real-time speech.

We are looking for founding researchers and engineers to solve hard problems in frontier generative speech and conversation models.

Remote·Full-time·Founding team

Open roles

Two founding seats. Both report directly to the founders and shape the core stack.

Founding Audio Researcher

Lead our work on novel audio LLMs and full-duplex audio modeling. You will design and train the next generation of our speech-to-speech models, moving beyond cascaded systems to end-to-end architectures.

The role

  • Research and train novel architectures for real-time speech-to-speech translation.
  • Explore full-duplex audio models that handle interruptions, backchanneling, and overlapping speech naturally.
  • Work on voice cloning and style transfer to preserve speaker identity across languages.
  • Read papers and implement state-of-the-art techniques quickly.

Requirements

  • Deep understanding of Transformer architectures, diffusion models, and neural audio generation.
  • Experience training large-scale models from scratch.
  • A product-first research mindset — you care about end-user experience, not just benchmarks.
Apply for this role

Founding Applied ML Engineer

Live at the intersection of model performance and system efficiency. You will own the inference stack, optimizing our models for low-latency real-time usage and on-device deployment.

The role

  • Optimize large audio and language models for production inference.
  • Implement efficient serving infrastructure using vLLM, TensorRT, or custom CUDA kernels.
  • Work on quantization and distillation for high-fidelity models on consumer hardware and mobile.
  • Handle model fine-tuning pipelines and data engineering.

Requirements

  • Strong engineering background with experience in low-level optimization (CUDA, C++).
  • Experience with inference optimization tools like TensorRT, vLLM, ONNX Runtime.
  • Familiarity with model fine-tuning (LoRA, QLoRA) and efficient training techniques.
  • Deep understanding of GPU architecture and memory management.
Apply for this role

About us

Pinch was founded on the belief that language should not be a barrier to human connection. We are a small, distributed team of researchers and engineers passionate about AI, audio, and design.

We operate remotely but maintain a high-bandwidth culture of collaboration, often traveling to co-work in person. If you're ready to move faster than any research lab you've worked at, we'd love to talk.

The Pinch team