Descrição da Vaga
We are a startup dedicated to transforming the legal profession and the delivery of justice through data.
We "read" and "interpret" massive case files of lawsuits being adjudicated in Civil (Consumer) courts in Brazil, and help our clients predict outcomes and define their optimal litigation strategies based on sound quantitative techniques.
Our clients are large corporations, mostly in the financial sector.
We are quickly expanding our team to increase our delivery capability to our clients.
Responsibilities You will be the chief person responsible for our algorithms and strategies to collect, structure and interpret data, which comes mostly as free-form text from PDF files written in Portuguese, but also as hybrid forms (images or tables merged with text).
You will take over our existing code base, and will have total freedom and responsibility to increase the breadth, accuracy and speed of our data mining, develop new predictive models for feature labeling and final case outcome, and for suggesting new use cases for our collected data.
Qualifications We are looking for a highly curious and knowledgeable person with mastery in NLP, ML and AI tools.
The successful candidate will have 3+ years of experience working with NLP, ML and AI, and will have mastered: Common mining strategies: simple and complex regular expressions, tokenization, lemmatization, text vectorization ML and AI algorithms for correct labeling Treatment of hybrid entities (images and tables merged with text) ML predictions for binary, continuous and time outcomes Tools/frameworks: Python, pandas, numpy, sklearn, TensorFlow, PyTorch, nltk, spacy, fitz, pdfPlumber, basic CloudOps Knowledge of Language Models is a plus.
Esta vaga foi obtida de um site de empregos.
