Software Development Generative Ai

Bharath Posa

940-***-**** ***********@*****.*** Github Linkedin

Summary

● Seasoned AI Architect with over a decade of expertise in designing and implementing Generative AI solutions.

● Proven track record across various levels in prestigious organizations such as JPMorgan & Chase Co. and Gap Inc.

● Holds a Master’s degree in Artificial Intelligence with a GPA of 4.0. Specialized in developing Retrieval Augmented Generation

(RAG) systems and fine-tuning large language models (LLMs).

● Extensive experience in all phases of software development, from ideation to deployment and testing. Over 11 years of experience in software development, consistently delivering innovative and high-impact AI solutions.

● Expertise in generating prompts and fine-tuning models for creating novel applications by applying prompt engineering techniques such as chain-of-thought, one-shot learning, general-to-specific, and tree of thoughts. Skilled in building applications using Retrieval Augmented Generation (RAG) and LangChain.

Professional Experience

Counduit LLC (Client AARP)

Senior Artificial Engineer Feb 2024 - July2024

Designed and developed a comprehensive chatbot solution for AARP Insurance using AWS Bedrock, AWS Lex, and AWS OpenSearch. Key contributions included:

● Architected and implemented the entire solution, ensuring seamless integration across web and mobile platforms.

● Created and managed knowledge bases to provide accurate and contextually relevant responses to user queries.

● Conducted extensive prompt engineering to optimize the chatbot’s natural language understanding and response generation.

● Integrated AWS OpenSearch to enhance the chatbot’s search capabilities, enabling efficient retrieval of information.

● Implemented robust security measures to protect sensitive user data and ensure compliance with industry standards.

● Performed rigorous testing and debugging wih EVAL metrics to ensure the chatbot’s reliability and performance under various scenario

Gap Inc

Staff Software Engineer Oct 2020 - July 2022

Gene Digital Assistant

● The Gene Chatbot/ Digital Assistant was developed to automate the tasks of the network team and answer questions in Natural language for management. It was deployed in the Microsoft bot framework connected in the backend by Stackato and deployed in Microsoft Teams.

● Achieved over 60% reduction in network operation turnaround time, significantly improving efficiency. Recognized as the winning solution in a hackathon, the application became an internal product, saving the company $250,000 annually and delivering interactive experiences in retail through Gap's Chatbot. Automation Framework 2.0

● Led a team in designing and implementing the Automation 2.0 Framework with a focus on eliminating repetitive tasks and enabling natural language processing for procedural network operations.

● Established a network of microservices to address diverse operational needs, ensuring scalability and efficiency in handling of complex tasks.

JPMorgan & Chase

Associate May 2018 - Sep 2020

● Created Data Pipelines to parse the issues related to the text of customer emails to chase credit complaints to classify them as forward to the appropriate escalation policy based on the severity

● Security compliance monitors provide a solution to automate the RHSA (Redhat Security Advisory) check with the Linux team on the latest security advisory by understanding the intent of the Security advisory and generating appropriate Jira Tickets. Vedams Software Solutions

Senior Software Engineer April 2014 - April 2018

Founded and spearheaded a team of 8 for machine learning and delivered an array of products in Machine Learning. Tulsi (OpenStack Swift Health Monitoring System)

● Tulsi (open stack swift health monitoring system) open source project for predictive maintenance of OpenStack swift project by knowing when drives go bad predictively using machine learning. Tulsi Book Chapter

Opioid Recommender

● Stanford opioid recommender helped in giving recommendations of the least addictive opioid drugs to patients in collaboration with Stanford University.

Jay Robotix Pvt Ltd

Research Associate Aug 2012 - March 2014 ● Worked on creating Innovative products related to child education and researched on sphinx software to create Text to Speech.

Defence Research Development Organization

Research Intern April 2009 – Aug 2011

Robotic Command System

● “Robotic Command System” is a self-ideated project that impressed and later got funded by the Govt of India to use robots to rescue people stuck in emergencies.

● This project made me put on different hats as a Designer, Robotic engineer, and Software developer. News Link

UNT

Masters in Artificial Intelligence Aug 2022 - Dec 2023

● Audio classification using Deep Neural Networks

Classify bird species using Sounds using librosa and custom CNN network models. This helped to classify the audio of birds on the Birdclef dataset.

● News Classification using NLP

Classify the degree of truthness in the news using XGBoost models. In this, we applied XGBoost to classify the text in Different degrees of truthness. This can classify text as partial truth, complete truth, or false.

● ML Service for prediction of Infrastructure resource utilization Forecasting system for predicting the Utilization of System Resources using ARIMA models implemented using AWS GLUE and lambda. In this, we created and deployed an ML service that has anomaly detection and Forecasting feature SKILL SET

Languages Python, Java

Natural Language Processing Natural Language Understanding, SpaCy,g, NLTK, Rasa, Prompt engineering, Microsoft Bot Framework, Azure cognitive services Large Language Models GPT3.5,BERT,PEGASUS, LSTM, T5, GPT 4.0, PaLM LLM Tools LangChain, LangSmith, LangServ

Amazon Web Service AWS Bedrock, AWS SageMaker, AWS IAM, CodeCommit, CodePipeline,Serverless (lambda), EC2, EKS, Secrete Manager, API Gateway, S3, RDS, Document base

Amazon Glue, AWS Glacier, AWS BFS,

Machine Learning Algorithms Convolution Neural Network (CNN), Artificial Neural Network, GAN, Deep learning, KNN, Supervised Learning, Unsupervised learning, Regression

Database MySql, NoSQL, MongoDB, Vector Database

Machine Learning platform Tensorflow, Pytorch, Pandas, Matplotlib, Numpy, Scikit-Learn, SciPy HuggingFace, Kaggle

Education

Master of Science in Artificial Intelligence, GPA 4.0 August 2022 - December 2023 Bachelor of Technology, Information Technology September 2007 - May 2011 Certifications

● AWS Cloud Practitioner

● OCI Generative AI professional

● OCI Datascience professional

● Microsoft AI Fundamentals AI-900

● Certified Cloud Foundry Developer

● Machine learning certification by Andrew NG, Coursera

● Data Science Foundation’s level 2, IBM

Contact this candidate