Shombeet Dasgupta

Generative AI practitioner on LLM based use cases like Semantic Search, vector DB, Question answering system using RAG and Text summarization.

Working on both Open Source or Paid LLM models.

Proficient in Python with indepth knowledge of NLP/ML.

Development of Conversational Chatbot to interact with various LLM's.

Prompt fine tuning to elicit information from LLM

Proficient in using langchain, llama index frameworks.

Applied diverse modelling techniques to address challenges within Natural Language processing and deep learning.

Integrated third-party APIs and components, expanding software functionality.

Designed and developed scalable software solutions, enhancing system performance.

Adept in using traditional machine learning algorithms

Produced high-quality model documentation for clear presentation to both technical and non-technical audiences.

Guaranteed timely and accurate model monitoring for ongoing adequacy and optimal performance.

Collaborated with team members to create applications' system analysis based on client requirements.

Developed software prototypes, enabling efficient testing and feedback.

Model development on Predictive Maintenance of Assets
Creating model to gauge customer sentiment analysis and presenting to stakeholders.
Extracting insights for business analysis using metering data and presenting to stakeholders.

Time series forecasting to predict yearly electricity consumption using Fbprophet.

Topic modelling of customer product reviews using LDA.

Applied diverse modelling techniques to address challenges within supervised and unsupervised algorithms(clustering, segmentation)
Worked in big data initiatives assisting on data lake creation with regards to MDM
Submitted dissertation for Masters using XAI(Lime and Shap) on hate speech detection and received top of the class distinction from LJMU as part of Masters in Data science and machine learning
Developed intricate algorithms based on deep-dive statistical analysis and predictive data modeling.

Created anomaly detection of customer consumption pattern using isolation forest.

Used business intelligence and other reporting tools to extract data from data solutions and data warehouses.
Analyzed large amounts of data to identify trends and find patterns, signals and hidden stories within data.

Working as a data analyst on oracle metering data management tools, working in analyzing customer billing data, usage cycles, creating visualizations in python on usage patterns

Activities include: Quality analysis, requirement gathering, deployment and UAT
Design of feature enhancements, reviews and integration
Created and automated data visualizations to present insights and tell compelling stories. Leading a team of developers

• Collaborated with team members to create applications' system analysis based on client requirements.
• Oversaw new feature updates to existing software systems to add functionality.
• Supported user-facing feature development and technical systems, maintaining reliability through monitoring and testing.
• Evaluated and selected new technologies to enhance software capabilities.
• Designed and developed scalable software solutions, enhancing system performance.
• Implemented debugging and troubleshooting techniques, enhancing software reliability.

Development of tools /adapters necessary for integration
Creating business level documentation for mapping from Utility/CIS system to AMI system.

Authored technical requirement documentation to suit business goals and technological limits.

Designed and maintained consistent data collection protocols and standards across numerous databases and projects.