Course code: WEB-047
  • Years with company: 1
  • Years programming: 5
  • Primary programming language: python
  • Other programming languages: r, julia, c++
  • Unit test harnesses: unittest, pytest, nose2
  • Something else: not much to say, i just gym, surf and do exercise besides work, i do alot of research into ml as its my academic focus
  • Test practice now: mainly ml code, so includes validating data pipelines, model performance, and reproducibility using the unit tests, integration tests, and metric based evaluations
  • Target system: microcontrollers, embedded devices, edge ai accelerators and npus
  • Dev tools: pytorch / tensorflow, huggingface, scikit, xgboost, mlflow, onnx, nvidia
  • Build time: 1-5 minutes
  • Coding standard: clean, modular, pep-8 compliant, type hint, config driven designs, and seperation of concerns
  • Function too long: too long when it tries to do more than one logical task, over 40-50 lines approx
  • Code reviews: focus on code clarity, reproducibility, data handling, evaluation logic. mix of software and ml
  • Code time: 50
  • Test time: 30
  • Debug time: 20
  • Favorite thing about dev: the freedom, and creativity you can explore with code, ml a mix of machine learning and standrd code
  • Least favorite thing about dev: sitting on my chair too long, back posture issues mainly
  • Tdd knowledge: test driven development is where you write tests before writing the actual code
  • Why are you attending: want to expand c++ testing as i dont do any, just python