I build production AI systems, RAG pipelines, and LLM apps trained by IBM, Stanford & DeepLearning.AI.
I'm Hamayl, an AI Engineer focused on building intelligent systems that combine computer vision, large language models, and modern AI architectures. My work centers on transforming AI research into reliable, production-ready solutions.
A structured, technical process for turning a research idea into a deployed, monitored AI system.
Scope the problem, the data available, and the constraints, latency, hardware, accuracy bar, before writing any model code.
Architect the pipeline, preprocessing, model choice, retrieval layer if needed, as a system, not a single notebook cell.
Train, fine-tune, and optimize for inference, augmentation, hyperparameter sweeps, and benchmarking against the constraint.
Ship to Streamlit, Gradio, or an API, then monitor, gather feedback, and iterate on the live system.
From real-time object detection to retrieval-augmented generation, each one built end-to-end, not a tutorial clone.
Computer-vision pipeline for autonomous drone-based infrastructure inspection, detecting defects and anomalies from aerial footage.
View on GitHubClinical Reasoning AI with RAG-Powered Medical Intelligence.A production-grade medical AI assistant with LLM reasoning to provide evidence-based clinical insights.
View on GitHubAutomated document understanding system, extracts, classifies, and structures information from unstructured PDFs and scans.
View on GitHubA multithreaded research-discovery assistant that helps users search, summarize, and explore academic literature through RAG.
View on GitHubLive video analytics system combining real-time object detection with downstream dashboards for tracking and insights.
View on GitHubA vision-language system built on BLIP that answers natural-language questions about the contents of an image.
View on GitHubSix specializations across machine learning, deep learning, and generative AI — completed end-to-end, not skimmed.
Complete 13-course specialization covering the full AI engineering pipeline.
Supervised, unsupervised learning, and advanced ML systems.
Neural networks, CNNs, RNNs, sequence models, and hyperparameter tuning.
Practical deep learning across the three leading frameworks.
LLMs, transformers, and generative architectures for NLP applications.
End-to-end data analysis pipelines, visualization, and statistical insight.
A real, working RAG-powered assistant trained on my own work and experience. Ask it anything about my projects, skills, or background, it's live right now.
Open to AI/ML engineering roles, freelance computer-vision and LLM work, or just a conversation about an idea you want to turn into a deployed system.