profile-pic

Gelli Tarun

Vetted Talent

Gelli Tarun

Vetted Talent
A skilled machine learning engineer passionate about solving real-world problems. Wish to explore this cutting-edge technology to help organizations develop new and integrate products Collaborated with multivariate teams of product development to insert trained models and gauge performance improvement. Planned, researched, and developed SOTA deep learning models to evaluate and perform semantic segmentation, object detection, and classifications. Developed data analysis and data preparation pipeline.
  • Role

    AI ML Engineer

  • Years of Experience

    3 years

Skillsets

  • Deep Learning
  • Gan's
  • Keras
  • LLMs
  • NLP
  • Numpy & pandas
  • Python Programming
  • Pytorch
  • R programming
  • Sciket-learn
  • Tensor flow

Vetted For

10Skills
  • Roles & Skills
  • Results
  • Details
  • icon-skill_image
    Machine Learning Scientist II (Places) - RemoteAI Screening
  • 66%
    icon-arrow-down
  • Skills assessed :Large POI Database, Text Embeddings Generation, ETL pipeline, LLM, Machine Learning Model, NLP, Problem Solving Attitude, Python, R, SQL
  • Score: 59/90

Professional Summary

3Years
  • Jan, 2021 - Present4 yr 3 months

    AI/ML AND DEEP LEARNING ENGINEER

    BIOMED INFORMATICS

Applications & Tools Known

  • icon-tool

    R PROGRAMMING

Work History

3Years

AI/ML AND DEEP LEARNING ENGINEER

BIOMED INFORMATICS
Jan, 2021 - Present4 yr 3 months
    Working as "AI Engineer and DATA SCIENTIST" Building AI models Built working models using DEEP LEARNING (Neural networks and ANN'S) Explaining the usefulness of the AI models to a wide range of individuals within the organization, including stakeholders and product managers Developing infrastructures for data transformation and ingestion Applied data science techniques, such as machine learning and statistical modeling

Achievements

  • Building AI models
  • Built working models using Deep Learning
  • Developing infrastructures for data transformation and ingestion
  • Applied data science techniques

Major Projects

24Projects

Detecting Diabetic Retinopathy

    Built a CNN model for detecting diabetic retinopathy and deployed it using TensorFlow Serving.

Stock Price Prediction Using DEEP-Q Learning

    Prepared an agent by implementing Deep Q-Learning that can perform unsupervised trading in stock trade. The aim of this project is to train an agent that uses Q-learning and neural networks to predict the profit or loss by building a model and implementing it on a dataset that is available for evaluation.

Health diseases Cardiovascular diseases

    Cardiovascular diseases are the leading cause of death globally. It is therefore necessary to identify the causes, so i had developed a system to predict heart attacks in an effective manner.

Stock Price Prediction Using DEEP Learning

    Prepared an agent by implementing Deep Learning that can perform unsupervised trading in stock trade. The aim of this project is to train an agent that uses Deep Learning and neural network models like RNNS AND LSTMS to predict the profit or loss by building a model and implementing it on a dataset that is available for evaluation.

Health Diabetic Retinopathy

    I had built a CNN model using distributed training that can detect diabetic retinopathy and deploy it using TensorFlow Serving.

Predictive Modeling

    I had built a machine learning model for a US Client which can predict runs of a batsman and number of wickets can be taken by a bowler in T20 matches using machine learning.

Health of Lung Infection

    I had built a model using a convolutional neural network that can classify lung infection in a person using medical imagery

Health diseases

    Cardiovascular diseases are the leading cause of death globally. It is therefore necessary to identify the causes, so i had developed a system to predict heart attacks in an effective manner

Chatbot Using Generative AI

    I had developed a real-time chatbot using LLMS and Layout LLM (Open ais Gpt-3, Whisper, Microsoft T-5) for sequencing and Whisper for speech to text processing to engage with the customers to boost their business growth by using NLP and Speech Recognition. We had deployed using Flask for web development & Microsoft Azure for deployment. The chatbot is very helpful for its 24/7 presence and ability to reply instantly.

Studio Looker

    This is a Solution for our Relationship Managers so that they can easily understand about our clients when they are about to contact and know about their likes and dislikes using the data aggregated of all features like from demographics to text to investments.

Top Up Model

    So, we had developed a model which uses the textual data from our clients to prioritize to contact our clients based on the probability by predictions, we used the meeting notes, call notes and the emails data and also with some feature engineered features from the above data.

Digital Footprint

    It is a dashboard created by us which uses the clients and prospects data from emails to calls to meetings counts and their digital activities and uses XGB Model to predict which client is low hanging fruit and fruitful to become a client or do a Top up if he is a client already. So, based on their activities to make it easy for our RMS to Contact and understand whom to contact in order.

Chatbot (SPEECH TO TEXT FOR CUSTOMER SUPPORT)

    I had developed a real-time chatbot to engage with the customers using voice commands and solving queries in order to boost their business growth by using NLP and Speech. The chatbot is very helpful for its 24/7 presence and ability to reply instantly.

Detection of Lung Infection

    Built a CNN model to classify lung infections in patients using medical imagery.

Health Care/Cardiovascular diseases

    Developed a system to predict heart attacks effectively, addressing a leading global cause of death.

Chatbot Development

    Developed a real-time chatbot with NLP and Speech Recognition to engage with customers and enhance business growth.

Facial Recognition

    Using a deep convolutional neural network (CNN) to perform facial recognition using Keras.

Emotion Recognition

    Future customizations, such as understanding human emotions, could lead to a range of advancements, such as determining whether a person likes a specific statement, item or product, food, or how they are feeling in a particular circumstance, and so on. I had built a model using a convolutional neural network that can classify a person's emotion

Lending Loan Data Analysis

    For some companies correctly predicting whether or not a loan will be a default and it is very important. In this project, using the historical data, I had built a deep learning model to predict the chance of default for future loans.

Prepared an agent by implementing Deep Q-Learning that can perform unsupervised trading in stock trade.

Stock Price Prediction

Health Informatics/Detecting Diabetic Retinopathy

Health Informatics/Detection of Lung Infection

Health Informatics/Cardiovascular diseases

Education

  • PGP IN AI/ML ENGINEER

    PURDUE UNIVERSITY, USA (2023)

  • B.TECH IN COMPUTER SCIENCE

    SREENIDHI INSTITUTE OF SCIENCE AND TECHNOLOGY (2022)

Certifications

  • Pgp in ai/ml engineer