Abhinav Shukla

Results-Driven Machine Learning and AI Expert with 7+ Years of Experience

Highly skilled and accomplished Machine Learning and AI professional with over 7 years of hands-on experience in the field. Adept at leveraging statistical analysis, predictive modeling, and artificial intelligence to drive data-driven insights and solutions. Proficient in both R and Python scripting, with expertise in Pandas, Scipy, Numpy, TensorFlow, and Keras.

• Machine Learning Expertise: Proven track record of creating descriptive, predictive, and forecasting models using both R and Python. Expertise in designing and implementing solutions for a wide range of analytical use cases.
• Deep Learning: Specialized in deep learning techniques, including RNNs (Recurrent Neural Networks), LSTMs, and CNNs. Proficient in using neural networks for text generation and computer vision projects.
• Big Data and Cloud: Skilled in harnessing the power of Big Data and Cloud technologies, including Hadoop, AWS (Redshift), and Snowflake. Experience in handling and analyzing large datasets.
• Text Analytics and NLP: Proficient in Text Analytics and Natural Language Processing (NLP). Capable of building models for sentiment analysis, text classification, and more.
• AI Solutions: Adept at creating and implementing AI solutions for various industries and problem domains. Passionate about exploring new analytical use cases and challenges.
• Programming Languages: Proficient in R, Python, Java, and experienced in deploying machine learning solutions on cloud platforms like AWS and GCP.

A dedicated problem solver with a strong commitment to staying at the forefront of AI and ML advancements. Excited to contribute expertise and innovation to new projects and challenges.

• Build works on Linode ( Cloud)
• Distributed queue system for Web
• Crawling
• Install worker documents supervisor Flower parallel-ssh
• Abhinav Shukla
• Data scientist AI/ML Engineer
• 07837153247
• Abhinavshukla92@yahoo.com
• Experience with Data science, Machine learning, Statistical modeling, Web crawling and Data mining
• Hopes to focus more on datascience and Artificial intelligence in future career
• I have over 7+ years of experience in Machine Learning(AI/ML), Statistical Analysis, Statistical Modeling, Artificial Intelligence
• Deep learning(RNN,LSTM,CNN),R Language, Python Scripting(Pandas, Scipy, Numpy, Tensorflow, Keras), Java ,Hadoop, AWSand GCP
• I have been working towards creation of descriptive, predictive as well as forecasting models using R and Python for multipleyears.Also worked on implementation of solutions using Big Data/Cloud technologies like Hadoop, AWS(Redshift), Snowflakeetc.
• Looking for exposure to new analytical use cases/problems
• I am also having experience in Text Analytics,Natural Language
• Processing, Artificial Neural Networks etc
• I have worked on solving problems using Neural networks like RNN(Recurrent Neural
• Networks,LSTMs) for text generation and have also worked on few personal computer vision projects as well
• Technical Corpus(Data Dictionary)
• Built a technical corpus/dictionary which will helppatent analysts to extract synonyms of technicalkeywords
• This dictionary was built using multipledata sources like Raw Patents brief summary text
• Wikipedia text and historical database queries
• Wordembedding models like Word2Vec were used to buildthis Corpus that contains around 1 billion words
• Text Summarizer
• Built a text summarizer system to provide briefsummary of the disclosure entered in the system
• Inother words the summarizer will act as a bot that willread patents for the analyst or customer and provide asummary for the same Extractive summarizationemphasized on summarizing the text by identifyingimportant sentences in the text
• This was done bysentence scoring i.e
• Giving sentences scores based onkeywords and key phrases present in them which wereidentified using Tf-idf, Textrank and Rake algorithms
• Top scoring sentences were selected to form thesummary.
• Patent Drafting
• The aim of this module was to generate first patentdraft from the idea given by an inventor
• In thismodule inventor can provide his idea in a fewsentences and the system will automatically generatepatent draft for the given text
• To accomplish this task
• LSTM neural network was trained on thousands ofpatent claims and descriptions to generate text fromgiven input
• Capital One: Data transformation
• Replacing old legacy data storage systems with
• Cloud storage(After attempting to move intohadoop ecosystem)
• The main objective of theproject is to move data from Teradata to AWScloud platform
• Data is migrated to S3 Lake the
• Object Storage service provided by AWS
• Datais also moved to Redshift -Data warehouseservice from AWS
• Before moving data to clouddata preprocessing is done
• Statistical modelingis done on the data after it is moved tocloud(Redshift)
• Expression Detection: Computer vision
• Worked on building a human facial expressiondetection web app using Deep Learning
• Convolutional Neural network model was trained onabout 35 thousand images with 7 differentexpressions classes(Angry, Disgust, Fear, Happy
• Sad, Surprise, Neutral).PIL and OpenCv were used toprocess images and capture real time video frameswhich worked as input for the deep learning model
• This app was deployed using Flask, Gunicorn and
• Nginx.