Deepesh Sharma

Data Scientist

Hyderabad, Telangana, India5 yrs 9 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Expert in developing machine learning models and frameworks.
Proven track record of optimizing data processes.
Strong leadership in data-driven project execution.

Stackforce AI infers this person is a Data Scientist with expertise in Fintech and advanced machine learning solutions.

Contact

Skills

Core Skills

Machine LearningData Science

Other Skills

AWS EC2AWS EMRArtificial Intelligence (AI)Autonomous ReceivablesC++CNNCluster AnalysisComputer VisionCore JavaData AnalysisData AnalyticsData WranglingDeep LearningExploratory Data AnalysisHugging Face

About

Deepesh Sharma is a skilled and experienced Data Scientist with a B.Tech. in Information Technology. With 5 years of experience creating machine learning models, statistical tests, predictive modeling, and data processing, He has contributed to developing a data science framework, implemented appropriate statistical techniques, algorithms, and ML models, and conducted full lifecycle analysis to optimize performance. Deepesh has supported the development of innovative business solutions by leveraging his domain knowledge, data science, and data engineering skills. Deepesh is passionate about exploring and implementing new data-driven techniques that optimize performance and drive growth. 𝗣𝗿𝗲𝗳𝗲𝗿𝗿𝗲𝗱 𝗽𝗿𝗼𝗴𝗿𝗮𝗺𝗺𝗶𝗻𝗴 𝗹𝗮𝗻𝗴𝘂𝗮𝗴𝗲: Python, C++ 𝗦𝗸𝗶𝗹𝗹𝘀: Machine Learning, Deep Learning, Predictive Analysis, SQL, PySpark, Exploratory Data Analysis, Data Wrangling & Creative Thinking. 𝗧𝗼𝗼𝗹𝘀: Jupyter Notebook, MySQL, Sckit-Learn, VScode, PyCharm, GIT , EMR ,EC2, Docker, Azkaban ,Pyspark ,Hugging Face Deepesh interest lies in building AI solutions and building customer-facing models and similar work as a Data Scientist/Machine Learning Engineer

Experience

5 yrs 9 mos

Total Experience

2 yrs 11 mos

Average Tenure

2 yrs 10 mos

Current Experience

Paytm

Senior Data Scientist

Aug 2023 – Present · 2 yrs 10 mos · Bengaluru, Karnataka, India

Paytm is India's leading digital commerce platform that offers services in Mobile Payments, Financial Services, E-Commerce, Lending, and Insurance.
𝐐𝐮𝐞𝐫𝐲 𝐀𝐠𝐞𝐧𝐭
𝟏) Built a Query Agent using Llama Index, Hugging Face, and Streamlit to convert natural language queries into Python/PySpark scripts for seamless data manipulation and extraction.
𝟐) Enabled business analysts to self-serve insights, reducing data team dependency and cutting turn around time from hours/days to minutes.
𝟑) 𝗥𝗲𝗱𝘂𝗰𝗲𝗱 𝗱𝗮𝘁𝗮 𝗿𝗲𝗾𝘂𝗲𝘀𝘁 𝗯𝗼𝘁𝘁𝗹𝗲𝗻𝗲𝗰𝗸𝘀 𝗯𝘆 𝟰𝟬%, accelerating user acquisition strategies.
𝐂𝐮𝐬𝐭𝐨𝐦𝐞𝐫 𝐒𝐞𝐠𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧
𝟏) Led a three-member team to deliver targeted customer segments across various Paytm businesses,
developing pipelines for periodic generation using NLP ,Hugging Face, PySpark , AWS EC2 and AWS EMR.
𝟐) Developed a Named Entity Recognition (NER) model using Roberta, achieving an F1 score of 0.94
𝟑) 𝗔𝗰𝗵𝗶𝗲𝘃𝗲𝗱 𝟰𝗫 𝗖𝗧𝗥 𝘂𝗽𝗹𝗶𝗳𝘁𝘀 𝗼𝗻 𝘁𝗮𝗿𝗴𝗲𝘁 𝗰𝗮𝗺𝗽𝗮𝗶𝗴𝗻𝘀.

PythonPySparkNLPHugging FaceStreamlitMachine Learning+1

Highradius

3 roles

Associate Software Engineer II (Data Science)

Promoted

Jul 2021 – Aug 2023 · 2 yrs 1 mo

HighRadius offers cloud-based Autonomous Software for the Office of the CFO. It brings modern digital transformation capabilities like AI, RPA, NLP, and Connected Workspaces for the finance & accounting domain.
𝐅𝐢𝐧𝐚𝐧𝐜𝐢𝐚𝐥 𝐃𝐨𝐜𝐮𝐦𝐞𝐧𝐭 𝐒𝐜𝐚𝐧𝐧𝐞𝐫
𝟏) A cutting-edge document scanner using CNN, YOLO, and OCR to extract financial data from large documents with precision.
𝟐) Worked on developing the pipeline to extract the relevant financial information ( ex : transactions , balance sheet, etc.)
𝟑) Reduced the 𝗱𝗼𝗰𝘂𝗺𝗲𝗻𝘁 𝗿𝗲𝘃𝗶𝗲𝘄 𝘁𝗶𝗺𝗲 𝗯𝘆 𝟵𝟬%, helping make decisions and actions faster.
𝐂𝐥𝐚𝐬𝐬𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧-𝐛𝐚𝐬𝐞𝐝 𝐫𝐞𝐜𝐨𝐦𝐦𝐞𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐬𝐲𝐬𝐭𝐞𝐦
𝟏) Built a Classification-based recommendation system to help our client's credit analyst make better decisions on block orders based on previous actions, payment information, historical orders, and
industry-specific credit information.
𝟐) Provided the reasons and recommended actions to help the credit analyst to make better decisions.
𝟑) The model helped in reducing the 𝘁𝗶𝗺𝗲 𝘁𝗮𝗸𝗲𝗻 𝘁𝗼 𝗿𝗲𝗹𝗲𝗮𝘀𝗲 𝗮 𝗯𝗹𝗼𝗰𝗸 𝗼𝗿𝗱𝗲𝗿 𝗯𝘆 𝘂𝗽 𝘁𝗼 𝟖𝟎% and also helped
in 𝗶𝗻𝗰𝗿𝗲𝗮𝘀𝗶𝗻𝗴 𝘁𝗵𝗲 𝗿𝗲𝘃𝗲𝗻𝘂𝗲 𝗯𝘆 𝘂𝗽 𝘁𝗼 𝟏𝟎-𝟏𝟓%

CNNYOLOOCRPythonMachine LearningData Science

Associate Software Engineer I (Data Science)

Sep 2020 – Jul 2021 · 10 mos

𝐂𝐮𝐬𝐭𝐨𝐦𝐞𝐫 𝐒𝐞𝐠𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧
𝟏) Worked on developing a customer segmentation model using K means clustering to segregate customers based on their payment patterns and the order amount.
𝟐) The model helped our client's collection agents to create a prioritized list of important customers to contact which led to on-time due payment thus 𝗶𝗻𝗰𝗿𝗲𝗮𝘀𝗶𝗻𝗴 𝘁𝗵𝗲𝗶𝗿 𝗿𝗲𝘃𝗲𝗻𝘂𝗲 𝗯𝘆 𝟐𝟓-𝟑𝟎%.

K-means ClusteringPythonMachine LearningData Science

Data Science Intern

May 2019 – Aug 2020 · 1 yr 3 mos

𝐂𝐚𝐬𝐡 𝐅𝐨𝐫𝐞𝐜𝐚𝐬𝐭𝐢𝐧𝐠
𝟏) Developed a time series model using Prophet Library to predict the upcoming cash flow for the time period of 1 to 6 months. 𝗢𝘂𝘁𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗲𝗱 𝗼𝘂𝗿 𝗰𝗹𝗶𝗲𝗻𝘁'𝘀 𝗜𝗻𝘁𝗲𝗿𝗻𝗮𝗹 𝗳𝗼𝗿𝗲𝗰𝗮𝘀𝘁𝗶𝗻𝗴 𝗼𝘃𝗲𝗿𝗮𝗹𝗹 𝗯𝘆 𝟓𝟑-𝟓𝟒%.
𝟐) Worked on Building a pipeline to automate the process for data collection, data preprocessing, and
predictions for the time series model.
𝐏𝐚𝐲𝐦𝐞𝐧𝐭 𝐃𝐚𝐭𝐞 𝐏𝐫𝐞𝐝𝐢𝐜𝐭𝐢𝐨𝐧
𝟏) Developed a regression-based model for the AR Collections System to predict the invoice payment
date of a given invoice in order to build a smarter collector work-list and proactive collection strategies for the AR Consultants.

Time Series AnalysisRegressionPythonMachine LearningData Science