Arindam Bhowmick

Project Management and Leadership:

• Architected enterprise-grade RAG pipelines on SAP BTP by integrating SAP HANA Cloud with vector databases and large language models, enabling context-aware intelligent search across business documents
• Led design and deployment of Agentic AI workflows in SAP S/4HANA and SAP SuccessFactors using SAP AI Core and SAP AI Launchpad, enabling autonomous decision-making for finance, HR, and supply chain operations
• Implemented GenAI use cases on SAP Cloud, including AI-powered ticket resolution, document summarization, and contract intelligence by orchestrating LangChain, OpenAI APIs, and SAP CAP (Cloud Application Programming)
• Collaborated with SAP product and data teams to enable secure consumption of LLMs in compliance with enterprise policies, while integrating fine-tuned models with SAP Business Technology Platform (BTP) for industry-specific solutions
• Evangelized RAG and Agentic AI strategy across enterprise architecture teams, setting up reusable GenAI accelerators, LLMOps pipelines, and governance frameworks aligned with SAP’s clean core principles

Design AI/ML Solutions:

• Design end-to-end AI/ML solutions that align with business objectives and technological capabilities
• Evaluate and select appropriate AI/ML frameworks, tools, LLMs and technologies
• Work with pre-trained models, awareness of state-of-art in embeddings and applicability for use cases
• Gen AI models (LLMs) like Mistral, Falcon, Llama 2, GPT 3.5 & 4, Prompt Engineering

Algorithm Development and Optimization:

• Implement advanced techniques such as deep learning, reinforcement learning, or natural language processing as needed
• Optimize algorithms for performance, scalability, and efficiency

Data Pipeline and Infrastructure Design:

• Design data pipelines for efficient data ingestion, processing, and storage
• Select and configure appropriate data storage solutions (databases, data lakes, etc.)

Model Deployment and Integration:

• Deploy machine learning models into production environments
• Integrate models with existing software systems and applications
• Monitor model performance and make necessary adjustments to ensure reliability and accuracy
• Ensure ethical and responsible AI practices are integrated throughout the development life cycle, considering fairness, transparency, privacy, and bias mitigation

• Developed and managed automation across ACCESSIBILITY domain
• Created an image classification model in order to reduce the time spent on testing different image parameters by 60%
• Optimized by automating the steps for identifying different image type and attributes resulting in 30% less time than previous releases
• Created an Energy Consumption Model in order to understand the areas fit for constructions and assist in selecting right design parameters
• Tech Stack Used: Data Science , Machine Learning, Deep Learning, Python, NLP
• Cloud : AWS, AZURE
• Libraries Used: Numpy, Pandas, Sklearn, Matplotlib, Seaborn, TensorFlow, Keras etc.

• Created an image classification model for aviation client in order to understand the areas where there is rust, oxidation loss, cracks in order to optimize the testing time for the product and assist towards safer travel
• Model brought down the time spent on testing by 40%
• Created a chatbot for a tourism client and deployed it over Slack in order to showcase the efficiency
• Have used libraries like NLTK, Spacy, Genism
• This resulted in addressing a lot of basic queries of the customers, which optimized the revenue of the client by 25%
• Created ML model using Linear Regression in order to address the shipping time delay and as well as to optimize the batch processing time by 30%
• Tech Stack Used: Data Science , Machine Learning, Chatbots using RASA NLU, Python, NLP, NLTK, Spacy, Bag of Words, Word2Vec
• Cloud: AWS, GCP
• Libraries Used: Numpy, Pandas, Sklearn, Matplotlib, Seaborn etc

• Designed and developed Machine Learning model to identify future job failures by using Decision Tree Algorithm
• Reduced the percentage of high and critical tickets, and optimized batch processing by 80%
• Created Flask Microservice in order to deploy the ML model locally, and as well as across different cloud platforms like GCP, AZURE and AWS
• Analyzed model performance based on different parameters and hyperparameter tuning
• Brainstormed ideas for client and helped in architecting applications using Machine Learning and Python
• Collected, studied, and interpreted large datasets; conducted reports; and visualization using Splunk

• Worked as Oracle Developer and Business Analyst
• Main responsibility is to resolve change requests, production deployment and sanity check, performance tuning of jobs
• Conducted CCB sessions in order to fetch new change requests
• Preparation HLD/LLD/TSD documents
• Development and unit testing
• Test case preparation and execution in ALM
• Test case reports preparation
• Demo of change with stakeholders
• Preparation of release documents
• BAT and UAT execution

• Test case preparation and execution in ALM
• Test case reports preparation
• Demo of change with stakeholders
• Preparation of release documents
• Sanity testing and smoke testing