We live in a moment of remarkable change and opportunity that is cognitive computing. The convergence of data and technology is transforming industries, society and the workplace. CognitiveScale is looking for talent to drive market success by building cognitive business solutions.
Our team integrates ML and NLP expertise with enterprise software to deliver innovative ML/AI solutions. This position is specifically for experts in the areas of data science, machine learning and statistics. We work in a dynamic environment where, depending on the complexity of the problem, we either use ML and NLP libraries from the public domain in a very heterogeneous set of platforms including Python, C/C++, and Scala, or design our own algorithms, adding to our core capabilities. These ML and NLP algorithms can require special attention to scale, careful utilization of data sparsity to promote faster execution, and deployment to cloud instances. Our task is to manage this complexity with quality engineering.
- Stellar background in statistical machine learning and data mining, in areas including but not limited to: online learning, nonlinear optimization, high-dimensional data clustering, classification, regression, ranking, probabilistic graphical models, and deep learning.
- PhD degree in AI, ML, statistical NLP or related technical fields.
- 10+ years of hands-on experience working in data science and/or machine learning.
- Practical experience building, testing and deploying ML models at scale.
- Experience with one or more of the programming languages and machine learning tools (Python, R, Julia, C/C++, Scala).
- Strong publication history in top-tier journals and conferences (e.g., Machine Learning Journal, DMKD, KAIS, ICML, KDD, ACL, NIPS, ECML, NAACL, IJCAI, AAAI).
- Strong interpersonal and communication skills and must be able to explain technical concepts and analysis implications clearly to a wide audience, including senior executives, and be able to translate business objectives into action.
- Experience with deep learning tools and packages (Caffe, PyTorch, Tensorflow).
- A clear passion for learning emerging technologies independently and continuously.
- Familiarity with models like LSTM, CNN, GCN, attention mechanism, GAN, VAE
- Experience with distributed training of deep models
- In-depth knowledge of probabilistic modeling and inference. Familiarity with models like Bayesian networks, undirected graphical models, implicit generative models and techniques like variational inference, MCMC sampling, large scale variational inference, stochastic gradient MCMC techniques is preferred.
- Knowledge of techniques like linear programming, convex optimization, cutting plane methods, momentum methods, stochastic gradient descent, alternating direction method of multipliers.
- In-depth knowledge of scalable inference in both directed and undirected graphical models should be very useful, given our heavy use of probabilistic graphical models to handle sparse, multi-relational and temporal data with complex interactions.