Data Scientist - NLP

Analytica•Washington, DC

19h•Remote

About The Position

Analytica is seeking a Data Scientist to support long term federal client engagements projects in the DC Metro area. The role will apply statistical programming, modeling, visualization techniques, data mining, and forecasting skills to analyze challenging public sector problems. This position is fully remote. This is a temporary position with the potential to convert to a full-time, regular employee role based on business needs, project funding, and individual performance. While full-time employment is not guaranteed, candidates who demonstrate strong performance and are a good long-term fit will be considered for permanent opportunities as they become available.

Requirements

Master's degree required, and PhD preferred in Statistics, Mathematics, Computer Science, or similar
High degree of experience utilizing SAS, R, or Python to support NLP use cases such as Document Summarization, Named Entity Recognition, Sentiment Analysis, and/or Topic Modeling
At least four years of experience developing scalable, production-ready NLP solutions using sci-kit learn, Keras, TensorFlow, PyTorch, Spark NLP.
Experience using git/github to version control source code
Experience leveraging transformer architecture to develop NLP models
Experience with open source NLP packages such as Gensim, SpaCy, or NLTK.
Experience with BERT, GPT-J, RoBERTa, T5 or other transformers
Experience with GenAI and Prompt Engineering is a plus
Experience in Databricks and MLFlow is a plus
Experience with machine translation and transcription of foreign language documents using Microsoft Azure translation services is a plus
Experience working in an AWS cloud environment and with related AWS services such as Bedrock and Textract
Experience coordinating and maintaining user stories
Must be a US citizen
Must be able to obtain and maintain a Public trust security clearance

Nice To Haves

PhD preferred in Statistics, Mathematics, Computer Science, or similar
Experience with GenAI and Prompt Engineering is a plus
Experience in Databricks and MLFlow is a plus
Experience with machine translation and transcription of foreign language documents using Microsoft Azure translation services is a plus
Experience working in an AWS cloud environment and with related AWS services such as Bedrock and Textract

Responsibilities

Pre-processing - Demonstrate the skills and experience to collect, clean, and prepare data sets for input into a computational model using Python. Strong candidates will explain various methods you have applied using common pre-processing functions such as stop word removal, stemming, lemmatization, and tokenization.
Feature Engineering and Attribute Evaluation - Candidate must demonstrate experience with NLP feature engineering methods such as TF-IDF, word2vec, GloVe, and FastText identifying the key determinants for modeling that exist in the business process and within existing data sets as well as selecting evaluation protocols (model techniques).
Modeling - Candidates will have practiced skills and experience selecting classification modeling techniques to fit the business problem. Examples will include techniques such as machine learning (ML) supervised and unsupervised learning, regression, neural networks and deep learning, natural language processing, etc.
Validation - Strong candidates will describe their experience with investigating, reporting, and justifying model results.
Visualization- Experience in presenting the results of their modeling activities, depicting the insights realized, and explaining the relevance of their results to the organization’s business challenges.