{
  "$schema": "https://raw.githubusercontent.com/jsonresume/resume-schema/v1.0.0/schema.json",
  "basics": {
    "name": "Dae-Young Jang",
    "label": "Robotics & ML Engineer, Forward Deployed Engineer at RLWRLD",
    "email": "djang@uwaterloo.ca",
    "url": "https://djang.dev/",
    "summary": "I build robots and the vision and learning systems that drive them. Right now that means humanoid teleoperation and robot learning at RLWRLD. Seeking a Winter 2027 co-op · Jan–Apr 2027, in robotics, computer vision or machine learning.",
    "location": {
      "city": "Waterloo",
      "region": "ON",
      "countryCode": "CA"
    },
    "profiles": [
      {
        "network": "GitHub",
        "username": "djang-c",
        "url": "https://github.com/djang-c"
      },
      {
        "network": "LinkedIn",
        "username": "dae-young-jang",
        "url": "https://linkedin.com/in/dae-young-jang"
      }
    ]
  },
  "work": [
    {
      "name": "RLWRLD",
      "position": "Forward Deployed Engineer",
      "location": "Bay Area",
      "startDate": "2026-09",
      "summary": "Teleoperating humanoid robots, then teaching them from the demos.",
      "highlights": [
        "Built a ROS 2 teleoperation pipeline that fuses a Quest headset, Vive tracker, Manus glove and two camera streams to drive an RBY1 + XHand robot, including the driver nodes and IK arm and hand retargeting.",
        "Ran teleoperation demos for AWS client proofs of concept that led to 3 strategic partnerships.",
        "Fine-tuned RLDX1 and RRC on 4 hours / 1,000 episodes of teleop data, reaching above 90% task success."
      ]
    },
    {
      "name": "RocScience",
      "position": "Computer Vision Engineer",
      "location": "Toronto, ON",
      "startDate": "2026-01",
      "summary": "Replacing a 30-minute sieve test with a real-time model.",
      "highlights": [
        "Cut soil particle-size analysis from 30 minutes to real time, validated against sieve testing, with a CNN + ViT ensemble trained on 12K+ images.",
        "Built YOLOv8 + SAM detection and segmentation that automated binary mask generation across incompatible datasets.",
        "Raised soil classification accuracy by 60% on unlabelled data with a ViT + Felzenszwalb + HDBSCAN clustering pipeline."
      ],
      "endDate": "2026-04"
    },
    {
      "name": "White Hat Consulting",
      "position": "Cybersecurity Engineer",
      "location": "Mexico City, Mexico",
      "startDate": "2025-06",
      "summary": "Automating the reporting so the team could do the security.",
      "highlights": [
        "Saved 16 hrs/month with a Power BI app on a private server, using REST APIs, Power Query and DAX to get past import limits securely.",
        "Saved 8 hrs/week by automating client report generation with a Python pipeline that builds and delivers Word reports."
      ],
      "endDate": "2025-08"
    }
  ],
  "education": [
    {
      "institution": "University of Waterloo",
      "area": "Mechatronics Engineering",
      "studyType": "BASc (Honours, Co-op)",
      "endDate": "2029-04"
    }
  ],
  "projects": [
    {
      "name": "WreckLess",
      "description": "An always-on security swarm on one desk-side Dell GB10: it reads a new advisory, exploits your code, patches it and proves the fix, then waits for a human to approve.",
      "keywords": [
        "Python",
        "vLLM",
        "Qwen3-Coder",
        "NVIDIA GB10",
        "Slack"
      ],
      "url": "https://github.com/djang-c/WreckLess-Cyber-Security",
      "entity": "Dell × NVIDIA AI Hackathon, San Francisco",
      "startDate": "2026-10",
      "highlights": [
        "3rd of 71 teams at the Dell × NVIDIA AI Hackathon, San Francisco, for running everything locally on the GB10",
        "Advisory to validated patch in 47–80 s: 5/5 live replays of CVE-2024-21513 with local models, no code sent to a cloud model",
        "Patch matched the fix upstream shipped (ast.literal_eval) without seeing it"
      ]
    },
    {
      "name": "Core Segmentation",
      "description": "Turns drill-core tray photos into one continuous, depth-labelled strip with every crack flagged.",
      "keywords": [
        "Python",
        "SAM 2",
        "Gemini",
        "MiDaS",
        "OpenCV"
      ],
      "url": "https://djang.dev/#p-core-segmentation",
      "entity": "RocScience core vision · own repo",
      "startDate": "2026-04",
      "highlights": [
        "20+ pixel-accurate masks per image across 100+ core samples",
        "I built the pipeline in my own repository during my RocScience term."
      ]
    },
    {
      "name": "The Flight of the Buffalo",
      "description": "An evidence engine for how rare diseases connect, with a lab-automation step that is checked in MuJoCo before anything runs.",
      "keywords": [
        "Python",
        "TypeScript",
        "MuJoCo",
        "Docker",
        "Cloud Run"
      ],
      "url": "https://github.com/djang-c/TFOTB",
      "entity": "Hack-Nation Challenge 05",
      "startDate": "2026-10",
      "highlights": [
        "52 source-quoted claims with DOIs · 2.48 s median for the five-call journey on the live app (10 runs)"
      ]
    },
    {
      "name": "Baseline",
      "description": "Live deepfake-voice and scam-call detection that listens to both sides of a phone call.",
      "keywords": [
        "Python",
        "TypeScript",
        "Whisper",
        "Ollama",
        "ElevenLabs"
      ],
      "url": "https://github.com/ANonABento/baseline",
      "entity": "Building from the Farm, SF Tech Week 2026",
      "startDate": "2026-10",
      "highlights": [
        "Warns at 0:21 in the scripted demo call and holds the transfer"
      ]
    },
    {
      "name": "WALDO",
      "description": "A pan-tilt camera that finds a person and turns to keep them in view, in real time.",
      "keywords": [
        "CAD (STEP)",
        "Python",
        "Alif E8"
      ],
      "url": "https://github.com/djang-c/Auto-Flashlight",
      "entity": "Own project",
      "highlights": [
        "Follows a moving person in real time on the built device (43 s recording)",
        "No part collisions across pan 0–180° and tilt ±90° (CAD clearance check)",
        "My own project: I designed the mount and wrote the clearance check."
      ]
    },
    {
      "name": "WATO ASD Robot",
      "description": "Point-to-point navigation for a simulated differential-drive robot: LiDAR costmap, A* planner and pure-pursuit control in a Dockerised ROS 2 and Gazebo stack.",
      "keywords": [
        "C++",
        "ROS 2",
        "Gazebo",
        "Docker",
        "Foxglove"
      ],
      "url": "https://github.com/djang-c/wato-asdrobot",
      "entity": "WATonomous ASD project",
      "startDate": "2025-09",
      "highlights": [
        "100% collision-free across the simulated test scenarios"
      ]
    },
    {
      "name": "Trash Can Robot",
      "description": "An ESP32-S3 robot that carries a trash can: the camera streams to a ROS 2 stack for inference and control, and the firmware drives two 12 V gearmotors.",
      "keywords": [
        "C++",
        "ESP32",
        "ROS 2",
        "OpenSCAD",
        "Python"
      ],
      "url": "https://github.com/ANonABento/trash-can-robot",
      "entity": "ESP32-S3 + ROS 2 build",
      "startDate": "2026-09"
    },
    {
      "name": "Segmentation by Patches",
      "description": "Segments very large images at full resolution: the image is tiled, a UNet with a swappable backbone segments each patch, and the masks are stitched back without downscaling.",
      "keywords": [
        "Python",
        "PyTorch",
        "UNet"
      ],
      "entity": "Own project",
      "startDate": "2026-01"
    },
    {
      "name": "Distance Scorer",
      "description": "Turns pixels-per-millimetre metadata into a calibrated apparent-distance score for soil photos, then trains a PyTorch regressor to predict it from the image alone.",
      "keywords": [
        "Python",
        "PyTorch",
        "SAM",
        "pytest"
      ],
      "url": "https://djang.dev/#p-distance-scorer",
      "entity": "Own project",
      "startDate": "2026-03",
      "highlights": [
        "Calibrated on 3,405 images spanning 8.8–136.2 PPM"
      ]
    },
    {
      "name": "EngSearch",
      "description": "A Claude Code research workspace: eleven specialist agents, with locked protocols, immutable raw data and run manifests enforced by hooks.",
      "keywords": [
        "Claude Code",
        "Python",
        "Shell"
      ],
      "entity": "Own project",
      "startDate": "2026-10",
      "highlights": [
        "Recovery/coverage 93–97% over 1,000 seeds in its validation log"
      ]
    },
    {
      "name": "GymBuddy",
      "description": "An AI-native iOS gym and nutrition coach: a deterministic, unit-tested Swift engine plus Claude meal parsing from text, voice and photos.",
      "keywords": [
        "Swift",
        "SwiftUI",
        "TypeScript",
        "Cloudflare Workers",
        "Claude"
      ],
      "url": "https://djang.dev/#p-gymbuddy",
      "entity": "Own project",
      "startDate": "2026-10"
    },
    {
      "name": "AppMaker",
      "description": "A multi-agent studio for designing and building iOS and web apps from audited, hash-locked skills. This site was built in it.",
      "keywords": [
        "Claude Code",
        "Swift",
        "Python",
        "Shell"
      ],
      "entity": "Own project",
      "startDate": "2026-10"
    }
  ],
  "skills": [
    {
      "name": "Robotics",
      "keywords": [
        "ROS 2",
        "C++",
        "IK retargeting",
        "Gazebo",
        "MuJoCo",
        "ESP32",
        "CAD (STEP)"
      ]
    },
    {
      "name": "Computer vision",
      "keywords": [
        "PyTorch",
        "ViT",
        "YOLOv8",
        "SAM / SAM 2",
        "MiDaS",
        "UNet",
        "OpenCV"
      ]
    },
    {
      "name": "Machine learning",
      "keywords": [
        "Robot policy fine-tuning (RLDX1, RRC)",
        "HDBSCAN clustering",
        "Monte Carlo",
        "Bayesian models"
      ]
    },
    {
      "name": "AI systems",
      "keywords": [
        "LLM pipelines (Gemini, Claude)",
        "claim verification",
        "multi-agent tooling",
        "voice deepfake detection"
      ]
    },
    {
      "name": "Software",
      "keywords": [
        "Python",
        "TypeScript",
        "Swift",
        "Docker",
        "Cloud Run",
        "Power BI / DAX"
      ]
    }
  ],
  "meta": {
    "canonical": "https://djang.dev/resume.json",
    "version": "v1.0.0",
    "lastModified": "2026-10-19"
  }
}