Skip to main content
Muhammad Anees

Muhammad Anees

AI/ML Data Annotator

Talent
IslamabadMember since November 20, 2026
PythonPyTorchTensorFlowKerasOpenCVLLMsAgentic AIn8nWhisperWhisperXLabelImgRoboflowMakeSenseCVATMQTTBounding Box CreationDataset CurationLabel StudioLabel DesignDoccanoRoboflows
·

Hire this person

Send a job offer directly to this candidate

PythonPyTorchTensorFlowKerasOpenCV

About

Electrical Electronic Engineering graduate with proven expertise in AI/ML data preparation and multi-modal annotation (text, images, and audio). Proficient in LabelImg, Roboflow, MakeSense, and CVAT for precise bounding box creation and dataset curation. Skilled in preparing high-quality training data for computer vision models (YOLOv5/v8) with strict adherence to project guidelines.

Seeking to leverage strong analytical skills and annotation expertise to ensure data accuracy and integrity as an AI/ML Data Annotator.

Experience

AI/ML Intern (Data Annotation Focus) at National Centre of Artificial Intelligence (NCAI), UET Peshawar (2026-10 – Present)

  • Annotated and labeled bacterial sample image datasets using standard annotation tools, ensuring high accuracy for AI research.
  • Collected, organized, and labeled media video data from multiple channels, maintaining data consistency and quality.
  • Followed strict annotation guidelines and identified inconsistencies in datasets to report to senior ML engineers.

AI Automation Intern at Patelnex Pvt Ltd, Pakistan (2025-10 – 2026-07)

  • Developed an automated video generation pipeline using Veo3, Kling, Runway, and FFmpeg.
  • Built an AI-powered podcast dubbing pipeline using WhisperX, Pyannote, and TTS APIs.
  • Developed a video-to-learning-content pipeline using Whisper, OCR, scene detection, and LLMs.

AI/ML Intern at IT Wings Technologies, Islamabad, Pakistan (2024-06 – 2024-11)

  • Developed a real-time Face Mask Detection system using a custom CNN (TensorFlow, Keras, OpenCV), achieving 95% accuracy on live video input.
  • Trained an EfficientNetB0 model via transfer learning for plant disease classification across 38 classes, reaching 94% accuracy.
  • Fine-tuned a VGG16 model for fine-grained classification of 120 dog breeds, reaching 69% accuracy.
  • Implemented real-time Multi-Object Tracking using YOLOv5 for detection and DeepSORT for identity association (PyTorch, OpenCV).

Education

BS in Electrical & Electronics Engineering – COMSATS University, Pakistan

Skills

AI/ML Data Annotator

Reviews

Related