Skillset:

  1. Candidate needs to explore the problem solving area independently without much of support as the work would be mostly in an exploratory and open-ended.
  2. Candidate needs to adapt to a new area of work specific to Media Pipe concepts like graphs, calculators, packets, synchronisation, real-time streams, on device GPU working, etc. Needs a bit of self learning skill.
  3. Candidate has to have a fair knowledge of integrating machine learning models that need to work within the mobile environment with constraints on no network, low memory and low compute. Especially with tensorflow and tensorflow lite. 
  4. [Good to have]   Candidate has to have a fair understanding of deep learning and model compression techniques like pruning, quantisation, sparsity of the model or low rank approximation, knowledge distillation, Neural architecture search. This would address the low resource environment constraint.
  5. Candidate has to have a fair understanding of image and video processing to address the preprocessing and post processing effects. 
  6. Prior experience in edge deployments or on device deployments.
  7. Candidate has to have experience with profilers, performance benchmarking.
  8. Candidate has to have experience with android and C++. iOS is a plus.
  9. Candidate has to have experience with handling 3D data for face mesh.
  10. Candidate has to have experience working with an SDK. 
  11. Candidate having a working experience with Augmented Reality is a plus
  12. Candidates having a working experience with ambient light sensors in mobiles is a plus.

Apply for this Job

* Required