Keep in touch with meI'm using Intch to connect with new people. Use this link to open chat with me via Intch app
Work Background
Wells FargoLeadership Reporting | Expert in Strategic Data-Driven Decisions
May. 2024New York, United States• emphasizing leadership reporting and encouraging data-driven executive decision-making. • overseeing and completing high-priority activities associated with deriving meaningful insights from data for GTM strategy, broader leadership goals, and digital projects for Commercial Banking & Corporate and Investment Banking (CB&CIB).
• Insightful, data-driven reports that highlight important metrics and trends related to CB&CIB digital projects and GTM strategies are developed and presented to senior leadership under supervision.
• Lead the process of extracting, analyzing, and interpreting data in order to spot patterns, opportunities, and problems.
• Stakeholder Coordination: Make sure data insights are successfully incorporated into decision-making processes by collaborating closely with cross-functional teams in Product, Marketing, and Technology.
• Advanced Analytics: Apply cutting-edge statistical techniques, machine learning frameworks, and predictive analytics to uncover deep insights from large and complex datasets.
Wells FargoPrincipal Data Scientist
Nov. 2023New York, United States• Pioneered the use of Spark and AWB for real-time analytics and insights. Collaborated with AI COE, DMI, and CB/CIB Data Management to improve business outcomes, reducing insight delivery time from hours to seconds.
• Patented a proposed Machine Learning/Topic clustering model, showcasing innovation in customer complaints management across two languages in CSBB.
• Served as the data science product owner/manager for two high-priority machine learning models, driving the project from conception to deployment.
• Improved project turnaround by 50% and achieved a 95% on-time completion rate. Analyzed server usage across 486 apps, leading to a $5 million saving for the firm.
• Coached and developed team members, offering mentoring on SAS, OSDS, AWB analytical Platform, H2O and data mining techniques, and scaled analytical capabilities across all business areas, influencing the bank's strategic planning and executive decision-making.
• Managed complex organizations developing, implementing, validating, and maintaining predictive models, including AI and ML
• Identified opportunities and strategies for analytics and data science, driving continuous advancement in the state of analytics across the enterprise.
• Developed and managed high-performance teams of quantitative analysts, fostering a culture of innovation and continuous improvement in mathematical and scientific disciplines.
Wells FargoCB&CIB Digital Analytics and ML Management - Commercial and Corporate & Investment Banking
Apr. 2022 - Mar. 2024New York City Metropolitan Area• Manage OKR Objectives & Key Results infrastructure
• Develop use cases & execute on ways to monetize CB & CIB Digital Data
• Analyze data looking for key trends to drive the business (leveraging machine learning techniques & unstructured data sets to drive behavioral & predicative analytics)
• Ensure Data & Product teams are utilizing the best technologies for clients & employees
• Integrate the insights from data and analytics for digital product owners
• Leading efforts to apply Advanced Analytics to get insights from customer data to help them with strategic advice and a comprehensive suite of corporate and investment banking services in CB&CIB
• Leading ML and FinTech efforts to provides market-leading solutions, industry expertise, and insights to help enable the clients' growth and success, enhancing the communities in CB&CIB
• Leading the design of a business model to identifying new growth opportunities with digital propositions to increase product-customer value in CB&CIB
• Leading the Digital Transformation in CB&CIB
Wells FargoProject Lead Developer
Jan. 2021 - Apr. 2022• Led and developed a Machine Learning model/product to predict the customer interest in the communication channels - Pending Patent
• Had the role of the data science and end-to-end product/pipeline manager for 2 high priority ML products for voice of customer complaints for ( English and Spanish) in Enterprise Customer Excellence Data & Analytic team.
• The owner, developer and product owner of a high priority project to identify the intention of customer’s call/communication. managed the product from end-to-end, handling the requirements of data ingestion and data pipeline, monitoring and controlling the model output and meeting the needs for monitoring reports for different stakeholders
• Productionalization of voice of customer Spanish Complaint model, testing and handing over the Model end-to-end from data ingestion to monitoring reports for different stakeholders
• Productionalization of voice of customer English Complaint model, testing and handing over the Model end-to-end from data ingestion to monitoring reports for different stakeholders
• Led and Engineered data pipeline for multiple model codes decreased the data ingestion and running time. One of the most successful one was decreasing time from average of 7~8 hours to less than 10 seconds
• Led the effort to adjacent product development and QA teams to successfully integrate four models/products for the Enterprise Customer Excellence Data & Analytic team.
• Pioneer and Lead of the Model testing, implementation proof document, and validation QA for end-to-end model delivery on production environment. This document has become the mandatory standard for the model delivery for all the sub-teams under the Enterprise Customer Excellence Data & Analytic Umbrella.
• The pioneer and lead of peer code review standards and documentation for all the sub-teams under the Enterprise Customer Excellence Data & Analytic Umbrella
Wells FargoSenior Data Scientist
Jun. 2019 - Jan. 2021Greater New York City Area• Proposed, developed and productionalized a new model for identifying the trending topics in voice of customer data based on time-pattern and BERT-CorEX topic modeling • Scaled analytical capabilities across all business areas, evolving analytics to influence bank's strategic planning and executives' decision-making by building 4 Machine Learning/NLP Models for voice of customer data - English and Spanish speaker customers
• Proposed, developed, and engineered preprocessing pipeline for voice of customer data which filters out the Spanish communications from English communications • Collaborated with internal stakeholders, identifying, and gathering analytical requirements for customer, product and projects needs.
• Tested and validated models for accuracy of predictions in outcomes of interest.
NANO Web Group | AI Venture StudioMachine Learning and Natural Language Processing Director
Mar. 2018 - May. 2019Greater New York City AreaAs a Lead ML and NLP engineer, I am responsible of applying the research and engineering skills to develop proprietary technology. I have several responsibilities such as: *Programming and Scripting Languages: Python *Database Systems: SQL (MySQL, SQLite), NoSQL(MongoDb), Postgresql, Flask *Frameworks and Toolkits : NLTK, SciPy, Numpy, Gensim, pytest, textblob, Stanford CoreNLP, Boost, igraph, Weka, Wordnet, Spacy, scikit-learn, pytorch *Problem Solving, Data Modeling and Analysis, Machine learning, Information Retrieval, NLP, Data Mining, Complex Networks Sampling techniques, Word Embeddings, *preprocessing data to transform and change raw feature vectors into a representation that is more suitable for our models *extract linguistic features like part-of-speech tags (POS), dependency labels and named entities (NER), *customising the stopwords, tokenizer *trained statistical models for spaCy named entity recognizer *creating rule-based matcher to find patterns *working with word vectors including tfidf, LSI,LSA, LDA *semantic similarity, sentiment analysis : Predicting similarity is useful for building recommendation systems and flagging duplicates. *clustering and classification :running a comparison of the clustering algorithms in scikit-learn to select the best model for the data
GoogleLead Text to Speech Language/Voice Engineer at Google
Feb. 2017 - Jan. 2018Text-To-Speech Voice/Language Engineer (TVS at Google) via Synergis
As a tts engineer at Google, I have several roles such as : 1. Improve synthesis in different languages. The aim in this role is to artificially produce human speech. I have to make sure that system converts text into speech under prescribed phonetic transcription of the specific language.
2. Developing rules for a text normalization. As language engineer I have to help improve the current algorithm or develop new one, to normalize a text and process it into a single canonical form that it might not have had before.
3. Large scale data mining and natural language processing. The role involves collecting and processing a large text corpus to train the automatic speech recognition system. By using NLP algorithms and tools, the language model is trained and it can recognize all domain-specific words and phrases.
4. Customizing language building tools for various languages. the tools and systems have to support new languages.
TreePressLead Data Engineer and Scientist
Aug. 2016 - Feb. 2017Greater New York City AreaIn TreePress, we are creating the world’s largest online network for theatre. A place where anyone, anywhere can upload, discover and explore scripts.
The scripts have different formatting and editing style. As a Data Scientist/Engineer creating this network includes different steps such as cleaning the data, detecting the different format of scripts, making a relational model of scripts, extracting the useful Information from Data and storing them in the database.
The first step is to distinguish and define different formats for the scripts. Based on those formats I have implemented a Parser Module which can detect the format of a script, extract the information from it and export it into database. The next step is to use machine learning algorithms to classify the scripts into different categories that we have, such as detecting the genre or style of a script. We are inspired by Convolutional Neural Networks, which rely on Text Classification Algorithms in NLP.
Later on, we aim to use interactive machine learning algorithms to make the extraction process more accurate. Writers know the input data(scripts) and its nature, therefor they can provide input to a learning algorithm, including input in the form of feedback, corrections and evaluations of the extraction process. Also, I have implemented approximate topic modelling to suggest subject matters for a document (script). Treepress is part of the Matter. group in New York City. . Matter supports media entrepreneurs building a more informed, connected, and empowered society through their start-up accelerator in San Francisco and New York City.
Intch is a Professional Networking App for the Future of WorkSee more people like Maryam on Intch
300k+ people
130+ countries
AI matching
Get Intch if you'd like to
489368 people
24
Data Scientist Intern @ Newell Brands
16
Program Manager @ DISH Network
18
Senior Software Engineer @ EarnIn
75970 people
24
Data Scientist Intern @ Newell Brands
15
Data Analyst
24
Looking to transition into Tech./. Former sales Representative and have very good Communication skilss @ @NikeStore