7+ years of academic and industry experience with 4.8 years of experience as Data Scientist using ML algorithms, Computer Vision and NLP.
Working experience as in building ML models using various algorithms and deep learning frameworks such as TensorFlow, Scikit Learn, and Keras.
Worked on tools like- PyCharm, Visual Studio, Jupyter Notebook.
Developed models for various applications, including computer vision, predictive analytics, natural language processing, anomaly detection, and recommendation engines.
Implemented a variety of optimization techniques such as cross-validation, hyper parameter tuning, and grid search to improve the accuracy of my models.
Working with GenAI capabilitiesvia textual and multimodal RAG pipeline
Experience in deploying models to production and have automated the process using technologies such as Docker, Kubernetes, and AWS.
Recognized and got appreciated by clients to deliver efficient solutions.
Have excellent communication and agile team working experience
Overview
12
12
years of professional experience
1
1
Certification
Work History
Data Scientist
IQZ Systems
Atlanta, GA
01.2022 - 02.2023
Designed an innovative algorithm that automated the resume review process, increasing productivity by 30%.
Involved in developing an LLM chatbot that can assist hiring managers in the resume screening process.
Involves the use of hybrid retrieval methods to augment the LLM agent with suitable resumes as context:
Similarity-based retrieval: When a job description is provided, the retriever utilizes RAG/RAG Fusion to search for similar resumes to narrow the pool of applicants to the most relevant profiles.
Keyword-based retrieval: When applicant information is provided (IDs), the retriever can also retrieve additional information about specified candidates.
The retrieved resumes are then used to augment the LLM generator, so it is conditioned on the data of the retrieved applicants.
The generator can then be used for further downstream tasks, like cross-comparisons, analysis, summarization, or decision-making.
Designed an embeddings-based remote application for a client to provide permission-based access to restricted areas.
Selected MTCNN for face detection, and FaceNet for embedding generation, utilized MongoDB as a feature store.
I used FastAPI as an interface for the model and checked it similarly using loss.
Implemented CI/CD in a monolith architecture using GitHub Actions, and deployed the web application on App Services on Azure Cloud.
R & D Engineer
Modern High Tech
Seoul, South Korea
09.2017 - 07.2018
Responsible for 2D object detection and classification projects, which include license plate detection, face and ID anonymization, vehicles, and person monitoring.
Requirement gathering and analyzing data of over 4,000 custom images.
Performed annotation and data engineering on the custom images.
Used OCR to fetch the number from the number plate detected.
Performed model selection and built a model on the images.
Achieved an accuracy of 86% after training the model on YOLOv4.
Deployed deep learning models as microservices using Docker, Flask/Nginx, and AWS.
Unit testing the data on customer datasets.
Created BI reports to present the data predictions.
Postdoctoral Research Fellow
Pusan National University
Busan, SOUTH KOREA
02.2017 - 08.2017
Engineered a fast, adaptable machine-learning model that improved the efficiency of solar cell accuracy by 30%.
Led a team of five to design and implement a predictive model using deep learning algorithms, increasing the performance of solar cells by 40%.
Developed a robust ML-based algorithm that reduced manual work by 50%.
Maintained accurate records of all fellows' progress throughout their tenure in the program.
Test Engineer
Nokia Pvt Ltd.
Bangalore, India
05.2011 - 02.2013
Worked on UMTS UE Protocol Conformance Testing (3GPP 34.123), with a focus on RRC, RLC, and MAC protocols.
Involved in GCF/PTCRB certification of Renesas chipsets.
Experience in testing the Renesas UE platforms on various network simulators, like Anite SAT6, Anritsu MD8480C, and Rohde and Schwarz (CMW500).
Education
Doctor of Philosophy (PhD) - Computer Science and Engineering
Pusan National University
Busan, South Korea
12.2017
Master of Technology (M.Tech) - Computer Science and Engineering
Sastra University
Tanjore, India
12.2012
Bachelor of Engineering(B.E) - Computer Science and Engineering
Certification on 'Online Summer Training Program/FDP on AI&ML and Blockchain Technology' through Finland Labs.
One Week Training On 'Image Processing Using OpenCV Python' offered through Finland Labs.
Obtained certification on 'Artificial Intelligence and Machine Learning' from Finland Labs.
Done certification course on 'Introduction to Machine Learning' through NPTEL.
Done certification course on 'Machine Learning' through Coursera.
Obtained certification on 'Resume Selection using Naive Bayes' using Coursera.
Obtained certificate on Online FDP on 'Data Sciences' from IIT Vadodara.
Projects
Feedback analysis based on sentiment review analysis on native language (telugu).
Detection Of Brain Tumor In MRI Images Using Anisotropic Filtering And Kernel SVM Classification
Computer vision related applications like Face Mask Detector in still images and video.
NLP related applications like Text Classification using Bidirectional LSTM.
Publications
Feedback analysis based on sentiment review analysis on native language (telugu), International Journal of Advanced Science and Technology Vol. 29, No. 5, (2020), pp. 11589-11596.
Two-Way Traceability using Smart Tags (QR): DL Tags, International Journal of Engineering and Advanced Technology (IJEAT) ISSN: 2249 – 8958, Volume-9 Issue-5, June 2020.
Detection Of Brain Tumor İn MRI Images Using Anisotropic Filtering And Kernel SVM Classification, International Journal of Advanced Science and Technology Vol. 29, No. 8, (2020), pp.3836-3844.