Sunanda Verma

Results-driven AI Engineer with over 11 years of experience in designing, developing, and deploying end-to-end machine learning, deep learning, and AI solutions. Proficient in Python, TensorFlow, and state-of-the-art NLP techniques. Hands-on expertise in building and fine-tuning Large Language Models (LLMs) using frameworks such as Hugging Face and Mistral AI. Specialized in predictive analytics, computer vision, and natural language processing (NLP), with a strong foundation in prompt engineering and generative AI using OpenAI's GPT-3.5 Turbo, GPT-4 Turbo, and GPT-4o. Demonstrated success in delivering scalable AI solutions by collaborating closely with product and engineering teams, optimizing model performance, and deploying models across cloud platforms like AWS, Azure, and GCP. Adept at translating business needs into intelligent systems that drive innovation, enhance decision-making, and deliver measurable impact.

Meticulous Data Scientist accomplished in compiling, transforming and analyzing complex information through software. Expert in machine learning and large dataset management. Demonstrated success in identifying relationships and building solutions to business problems.

• Nationality: Indian
• Driving License: UAE
• Marital Status: Married

• English, Professional
• Hindi, Native
• Urdu, Fluent

Personal AI-Based Projects

(Project links available upon request)

• Solar Panel Defect Analyzer – AI-Powered Solar Panel Defect Detection System

Automatic Computer vision solution for sustainable solar energy maintenance.

• Automatically detects manufacturing and operational defects in photovoltaic (PV) modules using advanced AI algorithms, including Convolution Neural Networks (CNNs) trained on Electroluminescence (EL) and Infrared (IR) images.

• Enables real-time classification of common issues such as micro-cracks, hotspots, and cell degradation — improving diagnostic speed and reliability compared to manual inspection.

• Provides automated recommendations for defect correction, grading panel health, and determining reusability potential, significantly optimizing maintenance schedules.

• Achieved up to 50% reduction in panel disposal through early fault detection and intelligent reuse mapping, supporting regional sustainability and circular economy targets.

• Designed to be user-friendly, allowing drag-and-drop image upload with instant AI analysis, making it accessible for both technical and non-technical users.

• Eco Track AI – AI-Powered Carbon Emission Footprint Tracker

An AI tool for individuals and organizations to measure and manage carbon emissions.

· Calculates carbon footprint based on energy use, transport, and lifestyle.

· Offers AI-generated personalized tips for footprint reduction.

· Visual progress tracking with intuitive charts.

· Provides real-time climate facts and insights.

• CO₂ Forecast System

Predicts future carbon emissions based on energy, transport, and lifestyle data to support sustainability planning.

· Developed a machine learning-based CO₂ footprint prediction system to forecast carbon emissions over the next 5 years.

· The model analyzes historical data related to energy consumption, transportation, water usage, and municipal services to generate accurate projections.

· Designed with interactive visualization features, the tool enables individuals and policymakers to assess future environmental impact and make data-driven decisions aligned with sustainability goals, particularly within the UAE and Middle East region.

• Weather Forecast System
Forecasts temperature and humidity trends for the next 10 years using historical climate data and machine learning.

· Developed a machine learning-based forecasting model to predict temperature and humidity trends for the next 10 years.

· The system uses historical climate data to train regression algorithms, enabling accurate long-term projections.

· This work supports environmental planning, sustainability analysis, and climate change research by providing insights into future atmospheric conditions.

• AI-Powered Domain Chatbots

NLP-based intelligent assistant for domain-specific support.

· Developed for medical and banking sectors to automate customer service.

· Reduced response time and eliminated waiting queues.

• Sentiment Analyzer – AI Sentiment Analysis Tool

NLP application for extracting insights from textual reviews.

· Analyzes customer feedback to identify sentiment trends.

· Helps businesses improve products and services based on user sentiment.

· Certified AI Engineer - London International Studies and Research Center (UK) (2025)

· Certified Data Science and Machine Learning Professional — London International Studies and Research Center (UK) (2025)

· PGDII with TUV Rhineland Germany Certified Qualification. PGDII with TUV Rhineland Germany Certified Qualification. (2023)

· Customized automation Customized automation SMEC COCHIN India (2023)

· PLC programming with Industrial Automation, KHDA Dubai (2023)

· Post Graduate Diploma in Industrial Instrumentation PGDII Post Graduate Diploma in Industrial Instrumentation PGDII (2023)

· Certification in Project Management, University of Adelaide, Australia (2016)

· Certification in Intellectual Property Law and Policy, University of Pennsylvania, Philadelphia, USA (2014)

· Certification in Solar Energy, Delft University of Technology, Netherland (2014)

1. Performance Evaluation of Different Medical Image Watermarking Techniques using EBCDIC coding presented in International Conference on Communication Technology ICCT 2013 and also published in proceedings IJCA‐ 2013.

2. Algorithm for evaluating false alarm probability of zero bit watermarking presented in National Conference on Computing, Communication & Information technology (CCIT 2010) organized by SLIET, longowal, Punjab.

3. Pixel level Document image processing for an OCR presented in National Conference on Microwave, Antenna and signal processing (NCMASP‐2011) organized by ITS Engineering college, Greater Noida.

4. Multiple Image Watermarking for Efficient Storage and Transmission of Medical Image published in journal and presented in conference ICACDS2019.

5. Contract Risks Identification and Mitigation in Solar Installation Projects Covering Prolonged Contract Duration published in journal and presented in conference NICMAR International Conference (ICCRIP–2018).

6. A Review Study of Detecting defects and failures of solar panels using image processing techniques published in journal and presented in conference International Journal of Engineering & Science Research (ICRIT 2018) ISSN No: 2277‐2685, (UGC Care).

7. Automatic Defect Classification of Electroluminescence Images of Photovoltaic Modules based on Deep Learning CNN published in International Journal of Mechanical Engineering, ISSN: 0974‐5823, Vol. 7 No. 1 January, 2022, (Scopus Indexed).

8. Defect detection of solar photovoltaic modules using inceptionV3 Deep CNN, NeuroQuantology, An Interdisciplinary Journal of Neuroscience and Quantum Physics, ISSN 1303 5150, (Scopus Indexed). · May 23, 2022.

9. Book Published with Title, Contracts Risk Management for Large Landscaping Contracts Involving Construction and Maintenance.

10. MEGAPROJECTS CONSTRUCTION MANAGEMENT IN VOLATILE MARKETS: Project Viability and Financial Sustainability Risks, Overcoming Climate Challenges and use of Artificial Intelligence.