Data Science Intern
AI / Research•Nairobi, Kenya (Hybrid / Remote)•Part-time / Full-time
You will work directly with our Principal AI Researcher on the models that power Gotham — Deeptrack’s enterprise deepfake detection engine. This means training runs, dataset curation, evaluation benchmarks, and performance analysis.
Responsibilities
- Run experiments on deepfake detection models trained against FaceForensics++ and real-world data
- Design and execute evaluation pipelines measuring precision, recall, and generalisation
- Curate, clean, and augment datasets for model retraining and fine-tuning
- Research and prototype improvements — temporal attention, multimodal signals, adversarial robustness
- Produce clear, reproducible research notes and model cards
Qualifications
- Strong Python skills; familiarity with PyTorch or TensorFlow
- Understanding of computer vision fundamentals and deep learning architectures (CNNs, ViTs)
- Experience with model evaluation metrics and statistical analysis
- Bonus: prior work with video/image forensics, GANs, or diffusion models
- Bonus: knowledge of C2PA or content provenance standards
- Currently pursuing or recently completed a degree in CS, Data Science, AI/ML, or related field
- Intellectually rigorous — you question assumptions and document your reasoning