Lakshmanarao Kota

2.5+years of experience as Data Scientist with a proven track record , specializing in AI-driven solutions. Expert in Python and analytical thinking, I developed models that reduced manual workload by 20%

Automatic Blog Posting Agent

One of the MNCs faced the problem of manual, time-consuming, and inconsistent blog posting
I addressed this by leveraging Agentic AI technology to automate the entire blog creation and publishing workflow.
Used CrewAI agents with OpenAI GPT-4o for topic research, content generation, and SEO optimization.
Resulted in over 80% reduction in manual effort, ensuring fast, high-quality, and consistent blog delivery.

• Healthcare users lacked an instant, reliable source for medical advice on diseases like AIDS, cancer, and diabetes, often relying on time-consuming, manual searches through complex medical references
• Developed an AI-powered medical chatbot that delivers real-time, accurate responses based on a medical textbook, improving accessibility to diagnostic, treatment, and condition-specific information.
• Parsed medical PDFs using PyPDF and embedded content using sentence-transformers (all-MiniLM-L6-v2), storing vectors in Pinecone for fast semantic retrieval.
• Integrated OpenAI LLM (llm = OpenAI(temperature=0.4, max_tokens=500)) with LangChain to generate concise, accurate responses and manage conversational logic
• Handled the complexity of long-form medical content, sparse keyword matches, and the need to generate medically coherent, contextually grounded responses using limited API tokens and controlled temperature settings

• Developed a deep learning model using VGG16-based transfer learning to detect early-stage chest cancer from CT scan images, addressing the challenges of manual diagnosis.
  Applied preprocessing techniques including resizing, normalization, and data augmentation to enhance model performance and generalization.
  Achieved robust results validated through metrics like accuracy, precision, recall, and confusion matrix.
  Deployed the trained model into a Flask web application for real-time CT scan analysis and prediction.