Summary
Overview
Work History
Education
Skills
Timeline
Generic

JUNZHENG WU

Toronto,ON

Summary

PhD-educated Senior Machine Learning Engineer driven by innovation and a passion for solving challenging problems with AI. Expertise spans conceiving novel frameworks (e.g., NL queries on SQL), developing bespoke algorithms, and building performant ML systems across diverse domains including Agentic AI, NLP, Computer Vision, and Knowledge Graphs. Consistently delivers creative and effective solutions, from initial research and design through to optimized deployment on platforms like GCP.

Overview

8
8
years of professional experience

Work History

Associate - Machine Learning

Dawson Partners
Toronto, ON
10.2023 - Current
  • Agentic AI System Development: Designed and deployed 3 end-to-end AI agent systems (GraphRAG, LangGraph) using FastAPI on GCP Cloud Run, serving 200+ internal users with CI/CD.
    Performance Agent: Enabled real-time financial insights by querying 30M+ Neo4j data, generating visualizations, and recalculating metrics.
    Document Agent: Engineered context-aware RAG for 80k+ documents, enabling complex queries with page-level citations.
    DDQ Agent: Optimized LangGraph state graphs and vector search to resolve historical due diligence queries in ~3s per round, supporting follow-up and session history.
  • Document Classification Pipeline: Developed scalable document classification pipeline (In-Context Learning, Pinecone) processing multi-hundred-page documents, generating structured summaries, and incorporating user feedback. Deployed on GCP Cloud Run (Pub/Sub) with outputs fueling the Document Agent.
  • Knowledge Graph ETL: Designed and implemented high-throughput KG ETL (Neo4j) unifying 20+ sources into 30M+ relationships from 30GB data in 1 hour using APOC optimizations and multiprocessing. Orchestrated daily incremental updates.
  • Semantic Search Engine: Delivered sub-2-second semantic search across 80k+ documents using vector search and Vertex AI Reranking.
  • Entity Matching System: Built real-time entity resolution system (

Deep Learning Scientist

GeneRxN (Generation) Research
Ottawa, ON
03.2020 - 10.2023
  • Designed an innovative framework for natural language queries on SQL databases leveraging LLMs (OpenAI, Langchain).
  • Developed deep learning models for medical image analysis: BiSeNet for brain stroke segmentation and high-accuracy EfficientNetB0 for Alzheimer's detection (TensorFlow, Numpy).
  • Orchestrated development of a live sensor data monitoring and visualization system.

Deep Learning Engineer

Neural Science Lab at University of Ottawa
Ottawa, ON
02.2019 - 02.2020
  • Developed and applied deep learning (TensorFlow) and computer vision techniques for neuroscience research.
    Implemented a time-distributed CNN for classifying mouse behaviors from video data.
    Pioneered image analysis software (DFT) and a novel algorithm (QuickNII) for medical/brain imaging.
  • Designed and implemented an automated system (Python/Arduino) for experiment control and data recording.
  • Contributed significantly to a research publication in eNeuro.

Deep Learning Engineer

Knowledge Vision
Chengdu
02.2017 - 09.2017
  • Company Overview: (Chengdu)
  • Fine-tuned a GoogLetNet Inception v3 model on a proprietary Invasive Ductal Carcinoma dataset and achieved an accuracy of 95%+ on the hold-out testing dataset using TensorFlow.
  • Skilled in deep learning and computer vision techniques.
  • (Chengdu)

Education

Doctor of Philosophy - PhD -

University of Ottawa
Ottawa
04.2022

Bachelor's Degree -

Southwest Jiaotong University
Chengdu
06.2018

Skills

  • LangGraph and GCP expertise
  • Deep learning and TensorFlow
  • Retrieval augmented generation
  • FastAPI backend engineering
  • Vector database management
  • Knowledge graph development
  • Python programming skills
  • Cypher query proficiency

Timeline

Associate - Machine Learning

Dawson Partners
10.2023 - Current

Deep Learning Scientist

GeneRxN (Generation) Research
03.2020 - 10.2023

Deep Learning Engineer

Neural Science Lab at University of Ottawa
02.2019 - 02.2020

Deep Learning Engineer

Knowledge Vision
02.2017 - 09.2017

Doctor of Philosophy - PhD -

University of Ottawa

Bachelor's Degree -

Southwest Jiaotong University

Similar Profiles

CREATE PROFILE
JUNZHENG WU