Open to opportunities

Michael Zhang

@michaelzhang1

Message

Senior machine learning engineer specializing in speech, LLMs, and multimodal AI.

Canada

Message

What I'm looking for

I seek roles building production-grade speech and LLM systems, emphasizing scalable architectures, real-time inference, cross-functional collaboration, and impact-driven product delivery.

I am a senior machine learning engineer with 8+ years building and deploying AI-driven systems from research prototypes to production-scale applications, with deep expertise in speech processing, LLM-powered automation, and multimodal pipelines.

I have engineered production voice restoration, voice cloning, and conversational voice agents integrating Whisper, Pyannote, RVC, Deepgram, ElevenLabs, Twilio, and GPT-series models; and built healthcare AI assistants leveraging PaddleOCR, LayoutLMv2, LangChain, RAG, and Claude/GPT models for clinician-ready outputs.

I design scalable, low-latency architectures using PyTorch/TensorFlow, FastAPI/Next.js, AWS, Docker, FAISS and real-time optimization techniques, consistently delivering improved reliability, accuracy, and user experience.

Experience

Work history, roles, and key accomplishments

Senior ML Engineer

Voxera Inc.

Sep 2020 - Nov 2024 (4 years 2 months)

Engineered a production voice restoration and conversational voice agent pipeline integrating RVC-based voice cloning, Pyannote diarization, Deepgram transcription, GPT-4o reasoning, ElevenLabs synthesis, and Twilio for real-time call handling, improving multi-speaker separation reliability and end-to-end conversational capability.

PyTorch Pyannote Deepgram GPT 4o ElevenLabs fastAPI React Faiss

Senior Machine Learning Engineer

VitalCore Health

Dec 2022 - Oct 2024 (1 year 10 months)

Designed and delivered an intelligent healthcare assistant and automated medical scribe platform using PaddleOCR, LayoutLMv2, custom Whisper transcription, LangChain RAG workflows and GPT-4o/Claude models, producing clinician-ready summaries and enterprise-grade chatbot capabilities.

Paddle OCR PaddleOCR LayoutLMv2 Whisper Lang Chain RAG GPT 4o Claude PyTorch Next.js

Senior ML / Speech Engineer

Voice.ai

Mar 2018 - Oct 2022 (4 years 7 months)

Developed real-time voice conversion and enhancement systems including fine-tuned RVC, VITS, Tacotron and HuBERT pipelines, neural noise reduction, and backend services for AI singing conversion deployed on AWS, improving clarity and latency.

Tacotron HuBERT AWS PyTorch