Machine Learning Engineer (Speech/Audio)

Company Overview

Plaud Inc., headquartered in San Francisco and incorporated in Delaware, develops a revolutionary AI platform aimed at enhancing human intelligence through seamless hardware-software integration. Since 2023, over 2.5 million users globally have benefited from its products that capture and structure intelligence from conversations to facilitate better thinking, quicker decision-making, and clearer execution. The company strictly adheres to stringent data security and privacy standards including ISO 27001, ISO 27701, SOC 2, GDPR, EN18031, and HIPAA.

Role and Responsibilities

• Take ownership of extensive speech and audio data processing pipelines, including tasks like data acquisition, cleansing, filtering, labeling, augmentation, and quality assurance to support machine learning model development.
• Collaborate with senior engineers on strategies such as hotword and term mining based on Automatic Speech Recognition (ASR) outputs.
• Assist in model training and fine-tuning of speech and language models, focusing on enhancing recognition accuracy for code-switching, proper nouns, and product-specific terms, while working closely with domain experts on algorithm design.
• Adapt models to specific domains using scenario-based data to improve accuracy for industry terms in key languages and verticals.
• Develop and maintain evaluation frameworks and test datasets, benchmarking Plaud's models against other open-source and commercial solutions using keyword and domain lexicon metrics.

Required Qualifications

• At least one year of practical experience in speech technology, machine learning, or managing large-scale data engineering projects.
• Proficiency in at least one of the following areas: training, fine-tuning, or evaluating ASR or SpeechLLM models; general LLM or ML model training/fine-tuning; or experience working with large audio, video, or text datasets at scale (e.g., tens of thousands of hours of audio or terabyte-scale multimodal data).
• Strong programming skills in Python and solid experience with PyTorch.
• Experience with distributed data processing frameworks such as Spark or Ray, especially given the responsibility of managing large-scale data pipelines.

Preferred Skills

• Knowledge of SpeechLLM or speech self-supervised learning concepts and familiarity with models like StepAudio or Qwen3-Omni.
• Prior involvement in developing or enhancing contextual biasing and hotword detection in ASR systems, including improvements for code-switching recognition.
• Academic publications or patents related to speech technologies presented at premier conferences such as Interspeech or ICASSP.
• Experience leading data pipelines involving extensive speech datasets in speech-model training projects.

Benefits and Culture

• Employee Stock Ownership Plan (ESOP) offering significant equity stakes and aligning employee interests with Plaud's long-term success.
• High-impact, fast-paced product environment enabling direct influence on the evolution of AI productivity tools.
• Access to leading-edge AI technologies including Gemini, Claude Code, among others, to maximize innovation and efficiency.
• Provision of premium work equipment including choice of top-tier laptops, high-performance workstations, and cutting-edge proprietary hardware devices.
• Engaging team culture featuring annual company off-sites, collaborative events, and an emphasis on craftsmanship, ownership, and rapid delivery.
• Comprehensive medical insurance and Work Injury Compensation Act (WICA) coverage to support employees' physical well-being and financial security.

Equal Opportunity Statement

Plaud Inc. is committed to fostering an inclusive workplace and does not discriminate against candidates based on race, religion, color, national origin, gender, sexual orientation, age, veteran status, disability, genetic information, or any other category protected by law.