AI/ML Experience:
Medical Reports Summarization (01/2025 to present):
- Designed microservices to generate medical summaries from medical reports using LLMs such as Grok 3, Gemini 2.5 Flash and ChatGPT4.0 Mini, to speed up medical insurance premium calculation by 70%.
- Libraries used: OlmOCR, OpenAI, Flask and Puppeteer.
- Technologies used: Python, Node.js, Firebase, DigitalOcean, Docker and Git.
Manufacturing Ticket Resolution Agentic AI Assistant (07/2024 to 12/2024):
- Designed and developed RAG-based AI agent to speed up customer incidents resolution by 40%.
- Libraries used: Langchain, ChatOpenAI, OpenAIEmbeddings, HanaDB, Streamlit, and FastAPI.
- Technologies used: Python, Databricks, Cloud Foundry, SAP HANA, SQL, Kubernetes, Docker and Git.
Customer Experience Improvement Through Emotion and Personality Trait Detection (01/2024 to 06/2024):
- Designed and implemented novel hybrid deep learning-based NLP models (CNN+bi-LSTM/CNN+bi-GRU) for improved text-based emotion and personality trait classification, resulting in better end-user product innovation and experience with 15% increase in sales.
- Libraries used: Flask, NLTK, Keras, Gensim (FastText, Word2Vec), GloVe, Sklearn and Tensorflow.
- Technologies used: Python, SQL, Kubernetes, Docker and Git.
Automated Ticket Classification and Routing for Banking Domain (03/2023 to 12/2023):
- Developed NLP based model, to automatically analyze and route product-related customer incidents to correct teams in real-time.
- Sped up incident resolution times by 20%.
- Python libraries used: Flask, Sklearn, Pandas, Wordcloud, NLTK and Spacy.
- Technologies used: Python, Databricks, SQL, Kubernetes and GIT.
Gesture-based Sentiment Analysis for Sales Domain (07/2022 to 02/2023):
- Developed deep learning-based machine learning model to capture customer sentiment based on real-time recognition of gestures, using pre-trained MobileNet model with GRU-based custom layers.
- Helped capture and communicate customer feedback to service providers faster, resulting in quicker remediation and increase in customer retention by 15%.
- Python libraries used: Flask, Tensorflow, Keras, CV2, SKImage, and Numpy.
- Technologies used: Python, SQL, Kubernetes and GIT.
Manufacturing Defect Detection (11/2021 to 06/2022):
- Implemented machine learning model (CNN+LSTM based model) for real time automatic visual product defect detection, during product manufacturing and shipping.
- Early defect detection reduced defective product delivery to customers to 0% and improved product quality and sales by 15%.
- Python libraries used: Flask, Tensorflow, Numpy, Pandas and Keras.
- Technologies used: Python, SQL, Kubernetes and GIT.
Logistics Delivery Prediction Feature (04/2021 to 10/2021):
- Built machine learning model using linear regression, to accurately predict delivery timings.
- This helped to implement dynamic routing mechanisms in real-time, thereby increasing on-time delivery of goods to customers by 25%.
- Python libraries used: Flask, Sklearn, Numpy, Pandas and Statsmodels.
- Technologies used: Python, SQL, Kubernetes and GIT.
Static Sales Forecasting for Automobile Industry (06/2020 to 03/2021):
- Developed machine learning model using linear regression, to better understand driving factors influencing sales and predict sales, thus improving overall sales through better risk mitigation.
- Increased overall automobile sales by 20%.
- Python libraries used: Flask, Sklearn, Numpy, Pandas and Statsmodels.
- Technologies used: Python, Flask, SQL, Kubernetes and GIT.
General Experience:
SAP Development Landscape Management - Feature Based Engineering (04/2025 - present) :
- End-to-end product design and development of checks in code pipelines of continuous feature-based enhancements delivered for the S4 HANA Public Cloud product.
- Reduced continuous feature enhancements' delivery times by 30%.
- Technologies worked with: Java, Python, OOABAP, Docker and Git.
SAP Digital Manufacturing Cloud - Central Initial Load, Delta Lake and Archiving (06/2024 - 03/2025) :
- End-to-end product design and development of archiving process, to replicate manufacturing data from analytics data store (warm store) to lakehouse (cold store) in SAP HDLFS, for building AI/ML capabilities upon.
- Technologies worked with: Python, Java Spring Boot, Docker, Kafka, Databricks, Kubernetes and Git.
SAP Digital Manufacturing Cloud - Insights (06/2023 - 05/2024) :
- Full stack development of new product features.
- Developed product features for real-time analytical views of manufacturing processes, that improved overall process efficiency of customers, by 15%.
- Technologies worked with: Java Spring Boot, Python, Docker, Kafka, SAP UI5, Tricentis Tosca, Kubernetes and Git.
SAP Business Networks – Supply Chain
Collaboration Monitoring (08/2022 - 05/2023) :
- Developed intuitive dashboards for seamless real-time business process monitoring experience.
- Helped reduce business issue resolution time by 20% via real-time status tracking of buyer/seller transactions.
- Technologies worked with: Grafana, Kapacitor, Telegraf, InfluxDB, Kafka, Docker, Kubernetes and Git.
SAP Internet of Things (IoT) – Tenant Onboarding (05/2021 - 08/2022) :
- Involved in new product feature development.
- Increased regressions caught during development and reduced manual test effort by 10% by increasing automation coverage.
- Technologies worked with: Node.js, Java, Python, Docker, Kubernetes and Git.
SAP Sales Cloud C4C/CX OData and REST API
framework (05/2017 - 04/2021) :
- Product design, development and customer incident handling.
- Improved code productive code performance by 40%.
- Technologies worked with: OOABAP, Node.js, Java, JMeter, Sahi Pro, Docker, Kubernetes and Git.