Your browser cookies must be enabled in order to apply for this job. Please contact support@jobscore.com if you need further instruction on how to do that.

Machine Learning Engineer

Engineering | Bangalore, India | Full Time

Job Description

About Us:

We are an early stage start-up working on next generation Digital Advertising platforms. We work on innovative approaches that connect brands, advertisers and publishers with consumers. We are looking at going beyond the current trends and hype to create a new standard in Digital Advertising. We have a stellar team plucked from great schools like Stanford. This is an opportunity to be associated with an early start-up team with a long term vision that also believes in having fun all the way. 

Responsibilities:

  • Work with massively dirty data
  • Develop prototypes, convert to production ready solution in a very quick time
  • Solve complex tasks in computer vision, Natural Language Processing but with simple Models that train and Infer
  • Build Data Wrangling, Analysis, Visualization and Modeling solutions on a laptop that deploy and scale with minimal change in code
  • Build model and data pipelines that scale over GPUs and systems on the cloud
  • Implement research papers flawlessly to adapt to newer datasets and replicate results
  • Experiment with models faster than they can train and use the scientific method to arrive upon a working solution
  • Understand Algorithmic complexity when working with data and ensure all development uses the most optimal solutions

Requirements:

We are looking for really smart, hardcore computer science engineers who can solve complex problems, munch through algorithms and deliver out of the box solutions.  Our problem domain is mainly computer vision.

Ideal candidates must have 1-2 years experience in the following:

  • Python/C++/R/Java
    • C++ - coding speed-up
    • R - statistics and plots
    • Java - Hadoop, mappers, reducers
  • Probability and statistics
    • Algorithms/models - Naive Bayes, Gaussian Mixture, Hidden Markov
    •  model evaluation metric - confusion matrices, receiver-operator curves, p-values, etc
  • Applied math and algorithms
    • SVM's
    • gradient decent, convex optimization, lagrange, quadratic programming, partial differential equations and alike
  • Distributed computing
  • Expertise in unix tools
  • Advanced signal processing techniques

Perks:

  • Early stage startup options
  • Create your own schedule