Skip to content
View maazmasood's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report maazmasood

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
maazmasood/README.md

AI Engineer — FullStack, LangChain, REST APIs

📞 +92-347-1531171 | 📧 maazmasood001@gmail.com | LinkedIn | GitHub | Upwork

Professional Summary 🌟

Experienced AI-Driven Full Stack Developer with over 4 years of hands-on experience in designing and delivering scalable web applications using technologies like Next.js, Node.js, PHP/Laravel, and React.js. Proficient in integrating AI capabilities such as intelligent automation and data-driven decision systems into modern web platforms. Strong background in both front-end and back-end development, and performance optimization. Known for building smart, user-centric solutions that enhance usability, accelerate workflows, and drive measurable business value. Actively seeking senior-level opportunities where AI and web technologies intersect to solve real-world problems.

Currently Working at 💼

AI Engineer — Full-time

MediRises LLC, USA (Remote)
July 2024 – Present

  • Architected and developed an AI-driven collaborative platform (inspired by Scribble) enabling real-time co-editing, drawing, and annotation using Next.js, Node.js, and LLaMA-based models.
  • Integrated advanced speech-to-text pipelines using Whisper for predictive typing, semantic suggestions, and intent extraction.
  • Engineered a low-latency real-time synchronization engine with WebSockets, supporting concurrent multi-user sessions with conflict resolution.
  • Built AI-powered autosave and summarization modules leveraging transformer-based architectures to generate concise overviews and preserve session state.
  • Enhanced system scalability by implementing asynchronous task queues and using SQLite as a lightweight storage layer for rapid prototyping and local persistence.
  • Developed structured clinical documentation modules with automatic generation of ICD-10 and CPT codes, improving medical billing accuracy and workflow efficiency.

Skills 🛠️

Languages & Frameworks

AI & ML

Front-End

Back-End & Databases

Tools & Platforms

Certifications 🎓

  • LangChain for LLM Application Development
    DeepLearning.AI
    Completed the "LangChain for LLM Application Development" course, instructed by Harrison Chase and Andrew Ng. This certification deepens expertise in building impactful AI applications by leveraging and chaining language model responses, a powerful technique shaping the future of AI development.

  • Finetuning Large Language Models
    DeepLearning.AI
    Completed the "Finetuning Large Language Models" course from DeepLearning.AI. This certification focuses on when to use finetuning versus prompting for LLMs, selecting suitable open-source models, and preparing data to train and evaluate models for specific domains—equipping me with the skills to optimize AI models for real-world applications.

Education 📚

Bachelor, Computer Science (2021–2025)
National University of Computing and Emerging Sciences, Islamabad, Pakistan
Extra Courses: Professional Practices in IT, Fundamentals of Marketing, Project Management

Pinned Loading

  1. Event-Management-System Event-Management-System Public

    A web-based platform for managing and organizing events at FAST University Islamabad. Features include user registration, event scheduling, participant management, multi-role access, and certificat…

    PHP 1

  2. LinkedinPostGenerator LinkedinPostGenerator Public

    JavaScript

  3. SecureQRCodeGenerator SecureQRCodeGenerator Public

    Python

  4. PSM-Anomaly-Detection PSM-Anomaly-Detection Public

    This project introduces GCT-GAN, a hybrid deep learning framework for unsupervised anomaly detection in multivariate time series. It combines Transformer Autoencoder reconstruction, GAN adversarial…

    Jupyter Notebook