<h2>Automating Golf Contract Audits with AI</h2> <h3>Faster Verification, Greater Accuracy, Less Manual Work</h3> &nbsp; <h3 class="MsoNormal">Introduction:</h3> Every year, golf courses across the country roll out limited-time promotional offers in an effort to draw in new players while also retaining their loyal customer base. As the season comes to a close, it becomes necessary to verify that the terms and pricing outlined in these promotional contracts align with the finalized pricing catalog. Traditionally, this auditing process was done manually requiring teams to sift through thousands of documents, which was not only time-consuming but also vulnerable to human error. This inefficient system prompted me to develop a smarter, automated solution. <h3 class="MsoNormal">Project Overview:</h3> Client Type: Internal marketing and operations department of a golf management company. &nbsp; Project Goal: To streamline and automate the end-to-end process of verifying submitted golf course contracts against the company&rsquo;s official pricing catalog. &nbsp; Tech Stack: Django for backend services, Next.js for the frontend dashboard, Playwright for automation, Redis for caching, and WebSockets for real-time communication. &nbsp; &nbsp; <h3 class="MsoNormal">Problem Statement:</h3> The manual auditing process was unsustainable and inefficient at scale. Reviewing each contract by comparing it line-by-line with the official pricing catalog required tremendous time and human resources. As a result, inconsistencies were often either overlooked or discovered too late in the workflow. This led to delays in approvals, potential revenue losses, and strained client relationships. It became evident that a scalable, intelligent system was needed to handle high volumes of contracts with precision and speed. &nbsp; <h3 class="MsoNormal">My Solution:</h3> I engineered a fully-automated, AI-enhanced platform that handles everything from the retrieval of golf course contracts to real-time comparison and reporting. The system integrates various modern technologies to ensure a seamless user experience while eliminating the majority of manual intervention. It was designed to be robust, scalable, and easily adaptable to other document auditing workflows in the future. <h3 class="MsoNormal">Development Process:</h3> 1. Smart Contract Collection:I implemented Playwright to systematically access various golf dashboards and download promotional contracts directly. This automation eliminated the need for manual login and navigation, significantly speeding up the data collection phase. 2. Data Extraction with AI:Using tailored AI-powered parsing logic, I extracted structured fields such as course name, pricing details, promotion dates, and terms directly from PDF files. This approach made the extracted data highly reliable and ready for downstream processing. &nbsp;3. JSON Comparison Engine:I created a specialized engine to compare the extracted contract data with the finalized catalog, highlighting mismatches in a clearly understandable JSON diff format. This made inconsistencies immediately visible to the end users. 4. Real-Time Frontend Streaming:Through the use of WebSockets, the platform streams live updates to the user dashboard, allowing them to observe processing in real-time and make decisions faster without needing to refresh the interface. &nbsp; <h3 class="MsoNormal">The Result</h3> Audit time was reduced from weeks to hours, freeing up valuable resources for other tasks. Accuracy was significantly improved, eliminating nearly all manual errors. Real-time insights helped stakeholders quickly approve or question contracts with confidence. The solution proved scalable and easily replicable across other documentation processes. <h3 class="MsoNormal">&nbsp;</h3> <h3 class="MsoNormal">Final Thoughts &amp; Takeaways</h3> Automating document reviews is not just about saving time it enhances operational trust and ensures decision-making is based on accurate data. The fusion of AI, real-time interfaces, and robust backend services can transform even the most cumbersome workflows into efficient processes. What once took weeks and a team of reviewers now takes only a few hours with minimal oversight. &nbsp; <h2>&nbsp;</h2>
<div dir="auto">Want to see this AI-powered golf contract auditing system in action?</div> <div dir="auto">Watch the video below to discover how it transforms slow, manual reviews into fast, accurate, real-time results.</div> <div dir="auto">&nbsp;</div> <div dir="auto">&nbsp;</div> <div dir="auto"><video poster="" controls="controls" width="300" height="150"> <source src="https://techtipsbymoh.tor1.cdn.digitaloceanspaces.com/media/project/teetime-project-passbook-automation.mov"></video></div>

Automate golf contract audits using AI. Compare documents in real time with higher accuracy, less manual work, and scalable backend technologies.

Tech Tips By Moh | Automating Golf Contract Audits with AI

Automating Golf Contract Audits with AI

A comprehensive guide to Artificial Intelligence — including its history, core concepts like machine learning and neural networks, real-world applications, ethical concerns, and future trends shaping technology and society.

Exploring the World of Artificial Intelligence

Learn how to use OpenAI API for developing AI-driven solutions, including creating projects like ChatGPT, with a focus on reusable and scalable coding practices using Django for backend development.

Tech Tips By Moh | Creating AI-driven Solutions with OpenAI API

Creating AI-driven Solutions with OpenAI API

Take your Docker skills to the next level in Part 2 of our guide. Learn how to write Dockerfiles, manage containers and images, use volumes and networking, and orchestrate multi-service apps with Docker Compose. Discover best practices to build production-ready applications with Docker.

Docker Explained: Advanced Concepts & Best Practices (Part 2)| Tech Tips By Moh

Docker Explained: Advanced Concepts & Best Practices (Part 2)

<h2>Live Progress Updates in Long-Running Backend Tasks: Polling vs WebSocket</h2> When your backend is handling long tasks like crunching big files, training a model, or working with AI APIs showing just a spinner until it&rsquo;s done is painful for users. A better way? Show live progress updates while the task is running, so users know something is happening. &nbsp; <h3>Why Live Updates Matter:</h3> Nobody likes waiting in the dark. Imagine uploading a 500-page PDF and staring at a loader for minutes. Instead, if the app shows progress updates after each page, users stay engaged and can even start using partial results before everything finishes. It makes the whole experience feel faster and more interactive. <h3>Method 1: Polling</h3> How it works The frontend keeps asking the backend every few seconds: &ldquo;What&rsquo;s the progress now?&rdquo; 1. User starts a task. 2. Backend processes the work. 3. Progress is stored (e.g., in Redis) with a unique ID. 4. Frontend hits a &ldquo;progress&rdquo; endpoint on repeat. 5. Backend returns the current status until the task is done. &nbsp; Pros: Super easy to implement. Works everywhere, no special setup. Cons: More network traffic (lots of repeated requests). Updates are tied to the polling interval, not truly real-time. &nbsp; <h3>Method 2: WebSocket</h3> How it works Instead of asking again and again, the frontend opens a WebSocket connection. From there, the backend can push updates instantly as they happen. Typical flow: 1. Frontend opens a WebSocket connection. 2. Backend runs the task and streams progress updates over that connection. 3. Frontend updates the UI immediately. Pros: Real-time updates with no delay. Less network overhead. Perfect for live dashboards, chat apps, and AI progress updates. Cons: More complex to set up and scale. Requires support on both frontend and backend. &nbsp; <h3>Practical Example:</h3> Here&rsquo;s a simple Python example to simulate progress: import time for i in range(101):&nbsp; # 0 to 100% &nbsp;&nbsp;&nbsp; time.sleep(0.2)&nbsp;&nbsp; # simulate work &nbsp;&nbsp;&nbsp; # send progress via cache (Polling) or WebSocket <ul> <li>With Polling, you&rsquo;d write progress to cache and let the frontend check it on intervals.</li> <li>With WebSocket, you&rsquo;d push progress directly to the client every step.</li> </ul> <h3>When to Use Polling vs WebSocket</h3> Polling is a good fit when: <ul> <li>Setting up WebSockets is hard in your environment. </li> <li>You don&rsquo;t need instant updates.</li> </ul> WebSocket is better when: <ul> <li>You want real-time, no-delay updates. </li> <li>You&rsquo;d like to cut down on repeated HTTP calls.</li> <li>Your app is interactive (chat, live feeds, dashboards).</li> </ul> <h3>Conclusion:</h3> Live progress updates change the way users experience your app. Whether it&rsquo;s AI, data crunching, or file handling showing progress in real time keeps users confident and engaged. Polling is simple and reliable; WebSocket is faster and more interactive. Pick the one that fits your setup best. Pro Tip: Using Django? Try Django Channels for WebSockets and Redis for progress tracking in both approaches.

Learn how to improve user experience in long-running backend tasks with live progress updates. Compare Polling vs WebSocket and choose the best real-time solution.

Live Progress Updates in Long-Running Backend Tasks: Polling vs WebSocket | Tech Tips By Moh

Live Progress Updates in Long-Running Backend Tasks: Polling vs WebSocket

<h2>Remove Image Background and Blend Images Using Python</h2>


 Looking to elevate your Python image processing skills? In this guide, you'll learn how to remove the background from an image using Python and blend it with another image using NumPy. It’s a simple yet powerful technique I recently shared in a video tutorial.


<h3>🎬 What This Tutorial Covers</h3>


 We'll walk through the process of combining two powerful Python libraries — rembg and NumPy — to create clean, professional-looking image composites. This is perfect for:


<ul>
 <li>Automating image post-processing workflows</li>
 <li>Creating thumbnails and social media visuals</li>
 <li>Building tools for AI/ML datasets</li>
 <li>Designing profile images or marketing content</li>
</ul>

<h3>📌 Step 1: Remove the Image Background with rembg</h3>


 The <a href="https://github.com/danielgatis/rembg" target="_blank" rel="noopener noreferrer">rembg Python library</a> lets you remove an image’s background using deep learning — no manual editing needed.


To remove a background:

<ol>
 <li>Install the rembg library using pip.</li>
 <li>Read the image file in binary mode.</li>
 <li>Pass the binary data to the remove function from rembg.</li>
 <li>Convert the output into an image and save it in PNG format.</li>
</ol>


 This results in a clean, background-free image that you can easily layer or reuse.


<h3>📌 Step 2: Blend Foreground and Background with NumPy</h3>


 Once you have the image without a background, you can blend it with a new background image using NumPy. Here’s how the process works:


<ol>
 <li>Resize both the foreground and background images to match in dimensions.</li>
 <li>Convert both images into NumPy arrays and use float values for blending.</li>
 <li>Apply a weighted sum using your preferred blend ratios (e.g., 80% foreground and 20% background).</li>
 <li>Convert the final array back into an image and save it as your final result.</li>
</ol>


 This lets you control how much of each image appears in the final result. You can experiment with different weights for creative effects.


<h3>📦 Full Code and GitHub Repository</h3>


 Want to try it yourself? I’ve published a complete working example you can clone and run. It includes sample images and instructions:



 👉 <a href="https://github.com/mmmohajer/image-overlay-and-grayscale" target="_blank" rel="noopener noreferrer">Image Background Removal and Blending – GitHub Repository</a>


<h3>🌟 Final Thoughts</h3>


 By combining rembg, Pillow, and NumPy, you can create powerful workflows for creative image editing, AI preprocessing, or content generation. Whether you're automating image cleanups or making dynamic visuals, this technique is a great place to start.



 If this tutorial was helpful, feel free to explore more on my GitHub or share this guide with fellow Python enthusiasts!

Learn how to remove image backgrounds using Python’s rembg library and blend images with NumPy for stunning effects. Explore the full code and try it yourself!

Removing Backgrounds and Blending Images with Python: A Simple Guide| Tech Tips By Moh

Removing Backgrounds and Blending Images with Python: A Simple Guide

If you&rsquo;re starting your journey in web development, you&rsquo;ve probably heard advice like this: &ldquo;Learn JavaScript, Python, and C++&mdash;you&rsquo;ll be a full-stack developer in no time!&rdquo; &ldquo;Master frontend and backend together&mdash;it&rsquo;s the fastest way to get a job!&rdquo; It sounds promising, but here&rsquo;s the truth: Trying to learn too many programming languages and frameworks at once is the fastest way to get stuck. <h2>Why Learning Too Many Languages Can Hurt Your Progress</h2> When new developers try to juggle multiple languages like JavaScript, Python, C++, and Java all at the same time, they spread their energy too thin. It&rsquo;s overwhelming, and instead of becoming a master in one area, they end up with surface-level knowledge that doesn&rsquo;t help them land a job or build real projects. This problem is so common that I made a video about it&mdash;watch it now to avoid falling into this trap! <h2>Why You Should Focus on One Language or Framework</h2> Here&rsquo;s the truth: ✅ Depth beats breadth. Learning one language deeply&mdash;like JavaScript with React for frontend or Python with Django for backend&mdash;helps you build real-world projects and solve real problems. ✅ Confidence comes from mastery. When you understand the core concepts of one stack, you&rsquo;ll find it much easier to pick up new tools later, like Docker, Redis, or CI/CD pipelines. ✅ Projects matter more than tutorials. Building a few solid projects teaches you far more than watching endless courses. <h2>How My Mentoring Can Help You</h2> If you feel overwhelmed by all the options, don&rsquo;t worry&mdash;you&rsquo;re not alone. In my mentoring program for aspiring web developers, I&rsquo;ll help you: <ul> <li> Choose the right tech stack based on your goals (e.g., frontend with React or backend with Django). </li> <li> Build hands-on projects that strengthen your skills. </li> <li> Stay focused on what really matters, so you can avoid getting distracted by the latest shiny tools. </li> </ul> The best part? The<a href="https://calendly.com/mohammad-aqd6/30min"> first mentoring session is free</a>&mdash;we&rsquo;ll explore your goals, your current skills, and create a clear, personalized plan for your learning journey. <hr> <h2>👩&zwj;💻 Ready to Start Your Web Development Journey the Right Way?</h2> 🎯 <a href="https://calendly.com/mohammad-aqd6/30min">Book your free mentoring session today</a> and let&rsquo;s build a roadmap for your success.

New web developers often struggle by learning too many programming languages at once. This blog explains why focusing on one language like JavaScript or Python can boost your skills and career faster.

Automating Golf Contract Audits with AI

Recent Blogs

Exploring the World of Artificial Intelligence

Creating AI-driven Solutions with OpenAI API

Docker Explained: Advanced Concepts & Best Practices (Part 2)

Recent Tips

Live Progress Updates in Long-Running Backend Tasks: Polling vs WebSocket

Removing Backgrounds and Blending Images with Python: A Simple Guide

The Biggest Mistake New Developers Make

Stay In Touch