Recruiter LogIn

Don't have an account? Register now

Forgot Password?

Recruiter Registration

Custom

Select Your Default Landing Page

Photo

SANTHOSH ***** AI ChatGPT Developer

$1140 / day

Summary

Senior AI Developer with over 15 years of IT experience, including significant expertise in Data Analysis, Generative AI, and team leadership. I specialize in designing and customizing generative AI solutions using technologies like DALL-E, Midjourney, and LLama (LLMs), integrating external data sources and automating testing frameworks. My skills span ETL processes, SQL, Python, AWS services, Power BI reporting, and DevOps practices. I excel in text analytics, Power BI development, and applying Agile methodologies to lead teams effectively.

SKills

Artificial Neural NetworksAWSAWS BedrockAzureAzure Data FactoryChatGPT APIDallE APIDatabricksDAXDecision TreesDeep Neural NetworksDevOpsDiffusion ModelsggplotGitHubJavaScriptJiraK-means ClusteringKerasKNNLarge Language ModelsLeonardo AILinear RegressionLlamaLogistic RegressionMatplotlibMidjourneyMS SQL ServerMySQLNLTKNumPyOpenCVPandasPHPPower BIPower QueryPower Query MPythonPyTorchRRandom ForestS3SageMakersci-kit-learnseabornSeleniumSnowFlakeSOAP UISQLSSASSSISSSRSSVMTalendTensorFlowVBA
AI ConsultantAI Software DeveloperMachine Learning Engineer

Education

2009 Master of Computers Application at Osmania University
2006 Bachelor of Science at Kakatiya University

Experience

Aug 2023 – present AI Project Lead at Zevoir Technologies

– Conducted data gathering, identifying the questions and responses that are frequently encountered in content data checks.
– Undertook exploratory analysis on customer queries to understand the primary topics and phrases.
– Leveraged Hugging Face’s Transformers library to convert unstructured data into entities and relationships, enabling efficient data organization and retrieval.
– Stored structured data in Neo4j and unstructured data vectors in Pinecone, attaching vectors to entities in Neo4j when necessary.
– Implemented a Retrieval-Augmented Generation (RAG) system by integrating a retrieval mechanism with the generative model to enhance the chatbot’s response accuracy using
external knowledge sources.
– Implemented the RAG system using Hugging Face’s tools to integrate structured data from Neo4j and vectorised unstructured data from Pinecone for accurate and contextually
relevant chatbot responses.
– Evaluated the chatbot’s performance based on metrics for content query resolution speed and accuracy.

Feb 2022 - July 2023 Senior Data Analyst / Scientist at CHEP

– I have designed technical architecture for the Data Warehouse.
– I have developed Machine Learning Models to predict the Demand using SageMaker and built the pipelines required (ML Ops) to automatically deploy train and deploy Models in AWS
environment.
– I have modelled the data tables and defined the required Dax Expressions for Reporting and Business Intelligence.
– I have developed, published, and scheduled Power BI reports as per the business requirements.
– Responsible for design methodology, security, and project documentation.
– Produce relevant statistics and visualizations to describe trends and patterns.
– I have documented technical specifications.
– I have integrated Power Apps into Power BI report and used Power Automate to create and run workflows from Power BI
– I have provided training for Basic Power BI users on advanced concepts and data modelling techniques.

Dec 2020 – Jun 2021 HSE Data Analyst (Data Scientist) at AGL

– Identifying the data and data source to be used for model building
– Exploratory analysis and creating visualizations, to get a better understanding of the data.
– We have used accuracy as the error metric to track the performance of different model and data combinations.
– We have used K-best, XG-Boost feature importance for feature selection to get the key features out of it.
– We have done K-fold cross-validation with five folds to check how the model will generalize
– We have used Support Vector Machines (SVM), Neural Networks, Random Forests and KNN for different combinations of data and hyperparameters and compared the results to find our
best performing model.
– Splitting the data by location and building a separate model for each location to predict incidents.
– We have used accuracy as the error metric to track the performance of different model and data combinations.
– We have done K-fold cross-validation with 5 folds to check how the model will generalise.

Jan 2016 - July 2017 Senior Engineer (Data Scientist) at United Health Group

- Preprocessing the data, we have used regular expressions to cleanse the data
– To get some insights into the data, we did an exploratory analysis and created visualizations using ggplot, matplotlib, word cloud and presented it to the client.
– We have generated features like “bag of words” and “TF-IDF” from the text.
– We have used K-best, XG-Boost feature importance for feature selection to get the key features out of it.
– We have used accuracy as the error metric to track the performance of different model and data combinations.
– We have done K-fold cross-validation with 5 folds to check how the model will generalize
– Presenting the Final Solution to the client and preparing documentation.

Jul 2009 - Dec 2012 Test Engineer at Infosys

– I was part of a team four member and was involved
– Designing a test automation framework in selenium and developing required libraries
– Automating test cases using Selenium (python), running and maintaining test scripts