Headline

Data Scientist | AI Engineer | NLP • LLMs • Generative AI | Python • FastAPI • Docker | Building Scalable ML Systems

About

I build end-to-end AI systems that deliver measurable business impact — combining Machine Learning, NLP, and Generative AI with strong engineering practices. During my internships, I’ve worked on solving real-world problems with clear results: >Improved customer churn prediction precision by 18% using XGBoost + SMOTE > Built JARVIS, an LLM-powered AI assistant, reducing manual task execution by 50% > Developed a real-time Neural Machine Translation system supporting 5+ languages, improving efficiency by 60% > Created time-series forecasting models with 91% directional accuracy > Automated data pipelines and dashboards, reducing manual effort by 30% --> What I Do I specialize in building production-ready ML pipelines: data → modeling → evaluation → deployment → monitoring --> Tech Stack Python • Machine Learning • Deep Learning • NLP • LLMs • Transformers • LangChain • RAG FastAPI • Flask • Docker • Streamlit • Power BI • SQL >> I’m passionate about building scalable AI solutions, especially in<< NLP & Conversational AI Generative AI applications Production ML systems (MLOps)

Location

Bengaluru

Work experience

Intern - Data Science & AI/ML at ExcelR, Apr 2025 - Nov 2025 · 8 mos, Bengaluru, Karnataka, India, Built a customer churn prediction model using XGBoost + SMOTE, improving precision by 18%, enabling proactive retention strategies Developed end-to-end ETL pipelines in Python, reducing manual processing effort by 30% Created 5+ interactive dashboards (Power BI, Plotly), saving 8+ hours/month in reporting

Data Science Intern at ExcelR Academy, Apr 2025 - Sep 2025 · 6 mos, Developed and deployed machine learning models (XGBoost, Random Forest) for customer churn prediction on 7K+ records, achieving 87%+ accuracy and enabling retention strategies. Built and optimized time-series forecasting models (LSTM, ARIMA) for stock prediction using RMSE, MAE, and hyperparameter tuning. Designed NLP-based chatbot systems using Hugging Face Transformers, GPT (LLMs), and Python, automating customer support workflows. Created Power BI dashboards and automated ETL pipelines using Pandas, Matplotlib, while applying featur engineering, model evaluation, cross-validation, Git, and MLOps practices.

AI/ML Intern – Rooman Technologies Pvt Ltd at Rooman Technologies Pvt Ltd, Jan 2025 - May 2025 · 5 mos, Bengaluru, Karnataka, India, Trained machine learning models for language translation and classification using scikit-learn and TensorFlow. Completed 480+ hours of AI/ML training under the NSDC Level 5 framework. Worked on neural machine translation (NMT) in collaboration with IBM Expert Labs. Earned certifications from Microsoft and IBM in generative AI and NLP technologies. Gained hands-on experience in data preprocessing, model evaluation, and real-world NLP projects. Developed practical skills in workplace ethics and communication through Wadhwani Foundation training.

AI/ML Intern at Rooman Technologies Pvt Ltd, Oct 2024 - May 2025 · 8 mos, Bidar, Karnataka, India, ▪ Designed and trained ML models for real-time language translation and classification ▪ Applied neural networks using Python, Scikit-learn, and TensorFlow ▪ Completed 480+ hours of structured NSDC Level 5 certified training

Education

Bachelor of Engineering - BE, Artificial Intelligence, Bheemanna Khandre Institute of Technology, Bhalki

Public links

github: https://github.com/viveksalimath