At a glance
- Experience
- 2-5 years experience
- Location
- Bangalore
About the role
- Glance is seeking an SDE II to work on Generative AI, focusing on model fine-tuning, optimization, and on-device deployment for mobile platforms.
- The role involves taking models beyond experimentation to ensure they function efficiently on consumer devices under constraints like latency, memory, and battery life.
Responsibilities
- Develop image generation, VTON, and identity-preserving generation models.
- Execute LoRA training, fine-tuning, and adapter-based personalization techniques.
- Conduct PyTorch-based training and experimentation.
- Manage dataset preparation, training pipelines, and model evaluation.
- Perform model conversion and deployment using ONNX and ONNX Runtime.
- Optimize inference using TensorRT and quantization techniques like FP16, INT8, and INT4.
- Implement on-device inference for iOS using Core ML and Android using TFLite/LiteRT.
- Profile and debug performance across CPU, GPU, and NPU/ANE.
- Reduce inference latency, peak memory usage, and model footprint.
- Build production-ready model pipelines from training through to device deployment.
Requirements
- 2-5 years of experience.
- Strong experience with Python, PyTorch, and deep learning.
- Hands-on experience with Computer Vision or Generative AI.
- Experience training or fine-tuning image-generation models.
- Understanding of LoRA, model optimization, and quantization.
- Knowledge of inference runtimes such as ONNX, TensorRT, Core ML, TFLite/LiteRT, or ONNX Runtime.
- Strong debugging, profiling, and software engineering fundamentals.
Skills
- Python
- PyTorch
- Generative AI
- Computer Vision
- LoRA
- ONNX
- TensorRT
- Core ML
- TFLite
- LiteRT
- FP16
- INT8
Show all 15 skills
- INT4
- CUDA
- C++
