AI Engineer - Data Science & ML Practitioner (Fresher)

Chinna Kasi Reddy Kandula

Mob: +91-901*******,

E-mail: **************@*****.*** linkedin.com/in/kasireddykandula github.com/Kandula0880

Objective:

To apply my technical and analytical skills in AI, Machine Learning, and Data Science to develop intelligent, real-world solutions that contribute to organizational growth Professional Summary:

Enthusiastic and results-driven Data Science & AI Engineer (Fresher) with strong technical expertise in Machine Learning, Deep Learning, and Computer Vision. Skilled in Python, PyTorch, and Scikit-learn, with hands-on experience in developing Generative AI and real-time object detection systems. Passionate about creating intelligent solutions that drive automation, analytics, and decision-making in real-world applications Education:

Bachelor of Technology (B.Tech) in Electronics and Communication Engineering Kalasalingam University 2021 - 2025 CGPA: 7.21 / 10 TECHNICAL SKILLS:

Programming & Databases: Python, SQL

Data Handling & EDA: NumPy, Pandas, Matplotlib, Seaborn Machine Learning Algorithms: Linear Regression, Logistic Regression, Decision Tree, Naïve Bayes, KNN, Random Forest, Gradient Boosting, XGBoost, K-Means, Hierarchical Clustering, Lasso, Ridge Regression

Deep Learning & AI: TensorFlow, PyTorch, YOLOv8, CLIP, VQGAN, LangChain, Mistral AI, Ollama

Advanced Concepts: Data Cleaning, Feature Engineering, Predictive Analytics, Model Evaluation, Model Deployment, Data Pipeline Design, Time Series Analysis, Anomaly Detection

Tools & Frameworks: Power BI, Streamlit, Gradio, Jupyter Notebook, PyCharm, Excel APIs & Deployment: Flask, REST API, Postman, Git, CI/CD, Docker (basic) NLP & Web Scraping: NLTK, Scrapy, Requests

LLMs & Vector Databases: GPT, LLaMA, Hugging Face, Embeddings, Fine-tuning Projects:

1. Text-to-Image Generation using VQGAN + CLIP :

Tech Stack:

Python, PyTorch, OpenAI CLIP, VQGAN, torchvision, NumPy, Matplotlib Description:

Developed a deep learning pipeline combining OpenAI CLIP and VQGAN to generate high-quality images from text prompts. Implemented optimization loops using cosine similarity loss to align image and text embeddings. Key Contributions:

- Integrated pretrained CLIP B/32) and taming-transformers (VQGAN) for multimodal learning.

- Designed custom normalization, transformation, and optimization modules.

- Tuned hyperparameters for stable convergence and realistic outputs. Outcome:

Generated meaningful visuals from natural language prompts, demonstrating AI creativity and multimodal learning.

2. Vehicle Counting & Speed Estimation using YOLOv8 + OpenCV: Tech Stack:

Python, OpenCV, YOLOv8, NumPy, PyTorch, Pandas, Matplotlib Description:

Built a real-time traffic monitoring system to detect, classify, and track vehicles using YOLOv8 and OpenCV. Implemented frame-based motion tracking for speed estimation. Key Contributions:

- Integrated YOLOv8 model for vehicle detection and classification.

- Used OpenCV centroid tracking for speed calculation using frame displacement.

- Applied ROI masking and perspective mapping for accuracy improvement.

- Optimized GPU inference to achieve 30+ FPS performance. Outcome:

Achieved 95% detection accuracy and reduced manual traffic monitoring efforts by 80%, supporting smart city safety analytics.

Certifications:

- Data Visualization - Forage (TATA)

- Data Analytics Job Simulation - Forage (Quantium)

- Data Analytics & Visualization - Forage (Accenture)

- Generative AI Studio - Simplilearn (Google Cloud) Strengths:

- Strong analytical and problem

-solving skills- Effective communication and presentation abilities

- Quick learner with adaptability to new tools and technologies

- Team player with leadership potential and organizational discipline Declaration:

I hereby declare that all the information provided above is true and correct to the best of my knowledge.

Contact this candidate