Lead Data Scientist - NLP/NLU/NLG & Speech - IISc/IIT/IIIT/NIT/BIT

Roles & Responsibilities:

- Architecting and Designing fast, scalable, and accurate NLP and/or Speech Processing based solutions at scale using open source and cloud based NLP and/or Speech Processing APIs.

- Engage customers as a thought leader by acting as the AI consultant / advisory especially in the area of NLP and/or Speech Processing.

- Lead and transform the NLP and/or Speech Processing stream in terms of innovation and solution building by creating differentiating solutions/features.

- Contribute as a leader to position Zensar in AI space, especially in NLP and/or Speech processing, by collaborating actively with other practices within Zensar.

- Building partnership with niche NLP and/or Speech processing solution vendors and using them to create our offerings.

- Engage in Presales solutioning sales enablement activities.

- Build and nurture a talented pool of NLP and/or Speech processing specialists in practice as a technical lead.

- Establish connects with industry stake holders.

- Contributing toward innovation by creating research work and patents in the domain of NLP/ML/DL, and Speech Processing.

- Able to translate state of the art research results into real world production systems.

Desirable Skills:

- 5-8 years of experience in building and architecting NLP/Speech solutions, majorly using Machine Learning and Deep Learning.

- Must have built production ready NLP/Speech processing systems using ML/DL.

- Experience in converting convert business problems to NLP/ML problems.

- Experience in building high precision fast and scalable NLP based system with little trading off in accuracy.

- Should have experience in AI consulting (especially in NLP & Speech Processing), customer interaction, building relationships, helping sell the deal

- Building high-impact proofs-of-concept to explore problem areas, then working with dev teams to turn them into real projects

- Should have a lead/mentored junior Data Scientists.

- Should have filed patents, written research papers& white papers in the area of NLP and/or Speech Processing.

- Should have a very good consulting experience - to understand the business problem, suggest the appropriate problems to be solved, and spearhead the initial execution.

- Very good hands on programming experience using Python, R, Java

- A deeper understanding of NLP-ML-DL and the ability to build models quickly using the appropriate framework.

- At least 4-8 years of hands-on experience in NLP using Statistical NLP, Semantic Web, Machine Learning (ML) and Deep Learning (DL) LSTM, GAN, CNN, GRU, ResNet

- Good experience in with building NLP application by creating/using word embedding (like Word2vec, Glov, FastText), auto regressive language models (like BERT, XLNet, ELMO), and topic models (like LDA, NMF).

- Should be hands on with Text Classification, Text/Document Similarity, Information Retrieval and document clustering.

- Should have worked on audio preprocessing, speech segmentation, speech recognition, and speech generations. Should be able to train and improve systems.

- Must have built production ready NLP/ML/DL based AI systems.

- Must have experience in creating and hosting scalable Restful APIs using R / Python.

- Must have the capability to convert business problems to NLP/ML problems.

- Understanding the right evaluation metrics for the NLP problems.

- CUDA based Deep Learning is highly desirable.

- Experience in distributes data processing using Spark especially Spark with MLlib

- Recruiting, hiring, team building, mentorship

- Should have experience in Customer interaction, building relationships, helping sell the deal

- Guiding and leading architecture of solutions.

- Building high-impact proofs-of-concept to explore problem areas, then working with dev teams to turn them into real projects

- Deeper understanding of Data Structure and Algorithms

- Should be able to read academic papers, follow the math, understand the connotations, and communicate technically with experts.

- A broad background in modern technologies and how they're used in production.

Good to have:

- Understanding of Recommender Systems (RS)

- Understanding of Predictive/Prescriptive Analytics like anomaly detection, fraud detection

- Understanding of Computer Vision theory and frameworks (like OpenCV, Dlib, SimpleCV, )

Mandatory Frameworks:

- R/Python/Java,

- Spacy, NLTK, Gensim, Core NLP, TextBlob, FastText

- Kaldi, DeepSpeech2, Wav2Letter++,

- Scikit-Learn, Spark MLlib,

- Keras, Tensorflow, MXNet,

- Hadoop, Spark,

- MongoDB, Cassandra, MySQL,

- Pandas, numpy, Scipy

- Spark

Qualification

- B.Tech/ M.Tech in Computer Science from IISc, IITs, IIITs, NITs, BITs or other reputed colleges. (5-8 years of relevant Industrial experience)

or

- Ph.D in Data Science from from IISc, IITs, IIITs (2-8 years of relevant Industrial experience post Ph.D). It is desirable that the Ph.D. is in NLP and Speech processing.