About me
I am a Data science engineer skilled in Python, SQL, and machine learning frameworks. Experienced in projects such as face recognition, indoor localization, and wake word detection, achieving measurable improvements like 10% accuracy gains on small datasets. Seeking to apply expertise to analyze data, build predictive models, and deploy AI solutions that deliver measurable impact.
I build end-to-end machine learning solutions, from data preprocessing and model development to deployment using modern frameworks. My work emphasizes accuracy, efficiency, and real-time performance, ensuring that AI systems are practical, scalable, and aligned with business needs.
What i'm doing
-
Machine Learning & AI
Designing and training machine learning models to solve real-world problems, with a focus on accuracy, efficiency, and practical deployment.
-
Computer Vision
Building computer vision systems such as face recognition and wake-word detection using deep learning models optimized for real-time and mobile environments.
-
Indoor Localization & Mapping
Developing WiFi-based indoor localization systems and navigation solutions using deep learning and pathfinding algorithms for accurate indoor positioning.
-
Model Deployment & APIs
Deploying machine learning models as scalable REST APIs using FastAPI, ONNX Runtime,TFLite, and cloud-ready architectures.
-
AI-Powered Web Applications
Integrating AI models into modern web applications using FastAPI and React to create intelligent, user-friendly systems.
-
Automation & Workflows
Automating data pipelines and AI workflows using n8n to connect models, APIs, and external services efficiently.
Technical Stack & Tools
-
Programing Languages
Python has been my go‑to for turning raw data into intelligent solutions, from machine learning models to real‑time systems. With libraries like NumPy, Pandas, TensorFlow, and PyTorch, I’ve built and deployed AI pipelines using FastAPI, ONNX, and TFLite.
JavaScript has been my bridge to building interactive web experiences, especially through React. At an intermediate level, I’ve used it to create dynamic interfaces and connect AI‑powered backends with responsive, real‑time applications. -
AI Frame Work
My journey with AI frameworks began when Python became a tool for building intelligent systems, using Scikit-learn and TensorFlow to solve real problems. I deploy these models into production by optimizing them with ONNX and TFLite and serving them through FastAPI for fast, scalable AI solutions.
-
AI Deployment
FastAPI is where my AI models become real products — I use it to turn trained models into scalable, high-performance APIs that can be reliably consumed by web and mobile applications.
Through FastAPI, I focus on clean architecture, low-latency inference, and seamless integration between AI backends and real-world systems. -
Automation
I use n8n to automate AI workflows and system integrations, turning repetitive processes into reliable, event-driven pipelines. Through automation, I connect models, APIs, and data sources to create end-to-end solutions that run efficiently with minimal manual intervention.
