Dr. Heeba Altaf Gurku

With over a decade of experience as a Doctor and a Machine Learning Engineer, I bring a unique blend of domain expertise and technical proficiency to drive AI innovation across industries. My background in oncology, general medicine, and AI/ML development has equipped me with the ability to build intelligent solutions that enhance decision-making, efficiency, and user experience. I specialize in Natural Language Processing (NLP), Large Language Models (LLMs), Prompt Engineering, Generative AI, and AI-driven automation, leveraging these technologies to develop impactful solutions. My experience spans clinical decision support systems, AI-powered applications, and predictive analytics, enabling me to bridge the gap between industry knowledge and advanced AI capabilities. By combining real-world expertise with technical acumen, I design and implement AI-driven solutions that optimize workflows, enhance user engagement, and drive innovation-delivering measurable improvements in outcomes and efficiency across diverse domains.

• Professional Data Science Certificate, IBM, Coursera
• Python 3 Programming Specialization, University of Michigan, Coursera
• Basic Statistics, University of Amsterdam, Coursera

Doctor AI - Healthcare Assistant  Developed an AI-powered virtual health assistant enabling seamless chat and voice interactions for  symptom analysis, medical report analysis and medical guidance. 

• Built Chat with AI & Talk to AI functionalities integrating AWS Polly (TTS) and  SpeechRecognition API (STT).  

• Implemented real-time AI responses via Claude 3.5 Haiku using Amazon Bedrock for medical  Q&A.  

• Developed a FastAPI backend, deployed with AWS Lambda & API Gateway for scalability.  

• Technologies: FastAPI, AWS Polly, Amazon Bedrock, SpeechRecognition API, Python,  Lambda, API Gateway.  

https://doctor-ai.drofn.com/  

Crypto AI (Talking Buddy) -Designed an AI-driven trading assistant for cryptocurrency markets with real-time market analytics. 

 • Developed predictive trading insights using LLMs, historical data, and technical indicators  (RSI, MACD, Bollinger Bands).

 • Integrated Binance APIs for live order book updates. 

 • Built a full-stack system with FastAPI (backend) and a responsive frontend using JavaScript.  

 • Technologies:  GPT-4o mini, Python, FastAPI, Binance APIs, Pandas  

 • https://cryptoai.drofn.com/  

AI-based Medical Appeal Generation System  Automated appeal generation for Physical, Occupational, and 

Speech Therapy claim denials using AI. 

 • Developed an LLM-powered system for analyzing denied claims and generating appeal letters. 

 • Reduced processing time by 50%, improving appeal success rate by 15%.

 • Technologies: Python, AWS, NLP, Large Language Models.  

Synthetic CT Image Generation using CycleGAN  Enhanced medical imaging for radiation therapy planning by generating synthetic CT images from  CBCT scans.

 • Developed a deep learning pipeline using CycleGAN & Attention-Guided GAN for accurate  image synthesis. 

 • Improved fidelity of synthetic CT images for better clinical applications. 

 • Technologies: Python, TensorFlow, PyTorch, CycleGAN, Attention-Guided GAN.