Asad Ali

🎉 Exciting News! 🎉 I am excited to announce that I have successfully finished my Airflow Fundamentals certification! This has improved my ability to coordinate intricate data pipelines and operations. I appreciate all of your help! #Certification #ApacheAirflow #DataEngineering #WorkflowAutomation

36

3 Comments

Excited to share that I've completed the Post Graduate Program in Artificial Intelligence and Machine Learning: Business Applications from The University of Texas at Austin - McCombs School of Business! 🎓 Grateful for the knowledge gained and excited to apply these AI/ML skills to future projects. #AI #MachineLearning #UTAustin #McCombs #LifelongLearning

25

10 Comments

𝗔𝗰𝗵𝗶𝗲𝘃𝗶𝗻𝗴 𝟵𝟴% 𝗿𝗲𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝗶𝗻 𝗲𝘅𝗲𝗰𝘂𝘁𝗶𝗼𝗻 𝘁𝗶𝗺𝗲 𝘂𝘀𝗶𝗻𝗴 𝗮𝘂𝘁𝗼𝗺𝗮𝘁𝗲𝗱, 𝗽𝗮𝗿𝗮𝗹𝗹𝗲𝗹 𝗰𝗼𝗻𝘁𝗮𝗶𝗻𝗲𝗿 𝘄𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀 𝗶𝗻 𝗔𝗪𝗦 While working for the University of Arizona's Department of Undergraduate Education, I was tasked with the containerization and cloud deployment of the analysis application I managed. While refactoring the application for containerization, I identified great potential for parallelization. This enabled me to set up an automated, parallel container execution pipeline in the AWS Cloud that helped reduce the analysis time for large datasets from a 𝘄𝗲𝗲𝗸 to a mere 𝗳𝗲𝘄 𝗵𝗼𝘂𝗿𝘀. Each application instance (container) performed analysis on a particular partition of the data. As the analysis of a partition was independent of other partitions, I recognized the possibility of executing all the tasks in parallel, significantly reducing execution time. For setting up and managing the automated execution workflow I made use of AWS Step Functions. Dedicated states in the workflow are responsible for autonomously generating the complex commands used for configuring the container task executions and scheduling tasks based on these. For cost effectiveness and minimal operational overhead, I decided to go with AWS ECS Fargate clusters to handle the provisioning and execution of the task containers. The final result was a 98% reduction in the analysis time. The analysis of the entire data now takes just around 𝟹̲–̲𝟺̲ ̲𝚑̲𝚛̲𝚜̲ rather than a 𝚠̲𝚎̲𝚎̲𝚔̲ like before. This not only helps speed up the analysis process but also expedites debugging and testing of new features/analysis profiles. Working on this problem gave me insight into distributed and parallel computing and application containerization in the AWS Cloud. I hope to find many more such learning opportunities during my tenure!

29

4 Comments

I’m excited to share that I’ve completed the Basic Data Descriptors, Statistical Distributions, and Application to Business Decisions course from Rice University on Coursera! 📊 This course enhanced my understanding of: Descriptive statistics and key data descriptors 📈 Statistical distributions and their applications in real-world business contexts 🔄 Leveraging data-driven insights to make informed business decisions and enhance strategies 💼 #DataScience #Statistics #BusinessDecisions #DataDriven #RiceUniversity #Coursera #DescriptiveStatistics #StatisticalAnalysis #LifelongLearning #BusinessIntelligence

22

1 Comment

I'm excited to share that I've published an article on Medium summarizing Lecture 4 of the "𝗗𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗮𝗻𝗱 𝗶𝘁𝘀 𝗔𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀" course taught by Dr. Muhammad Attique Khan in my 6th semester. In this article, I cover: • What are Hyperparameters? • Why Do We Use Hyperparameters? • What’s the Role of Hyperparameters? • Types of Hyperparameters • Examples of Hyperparameters • What is Gradient Descent? • The Math Behind Gradient Descent • Types of Gradient Descent • Convergence and Learning Rate Special thanks to Dr. Muhammad Attique Khan for his insightful teaching! Read my article on Medium:

36

3 Comments

I am humbled to be invited to speak tomorrow at Rice University in Houston, TX for the IEEE Try Engineering Summer Institute. My session will cover my current position, career path, challenges faced and how they were overcome with advice on how to connect with others in industry. A few key points from my session: "Embrace change, don't fear it" "Focus on what is possible, not on what is impossible" "Identify your goals, never stop learning and progressing toward them" "Always look for opportunities to expand your network and ask what you can do to further their goals as well as your own" "Don't wait to be discovered, go discover for yourself" I will cover my entry into industry as a coder, how I faced my fears, overcame obstacles, formed relationships, grew my network, learned to lead from my mentors and conclude with my current path as an advisor to companies providing analysis, recommendations, and helping create paths to achieving their own success. Along the way I got to experience: Flying helicopters Starring in my own TV commercial Being a "Stand-In" for a movie star Learning to fly the Space Shuttle Having lunch in the center of the Pentagon "The best way to test whether a person can lead rather than just manage is to ask them to create positive change" - John C. Maxwell, The 21 Irrefutable Laws of Leadership

42

10 Comments

Zeeshan Usmani, Ph.D is currently on a tour in Pakistan, and I've had the opportunity to listen to 3 to 4 podcasts from his recent tour. Among them, the one with Out of Sync Podcast hosted by Devsinc issues stands out as particularly great. Hosted by Fahd Faheem, the questions posed were tailored to address the current needs for guidance and planning among the youth. As per insights shared by Zeeshan Usmani, Ph.D, one of the world's leading data scientists, there are three key rules to incorporate into your life if you aim to achieve success: Discipline, Focus, and Consistency. Adhering to these principles can pave the path to success. 1. **Discipline**: Managing daily tasks is crucial. Whether it's waking up at 5 am or having meals at 6 pm, sticking to a schedule is essential. 2. **Consistency**: It's not about doing more; it's about doing consistently. 3. **Focus**: Multitasking can be detrimental to our productivity. Engage in one task at a time, giving it your undivided attention. Whether it's reading a book or writing code, keep distractions like mobile phones at bay. For software engineers and other professionals, Dr. Usmani advises everyone to prioritize reading. With rapid advancements and developments occurring globally, staying updated is imperative. Employers expect their workforce to remain abreast of the latest trends and technologies. #Success #CareerAdvice #Discipline #Focus #Consistency #ContinuousLearning #Engineering #ZeeshanUsmani #Devsinc #SoftwareEngineer #DataScientist #Technology

30

4 Comments

🔒 Excited to share my latest project, SecureVault, developed for my Cyber Security course at CU Denver College of Engineering, Design, and Computing under the guidance of Professor J. Haadi Jafarian. SecureVault is a password-secured vault for Android devices that operates solely during specified hours and GPS locations. This integration of time and GPS-based access control enhances security, ensuring sensitive data is only accessible when and where it is safe. 💻 Developed using Java, SecureVault utilizes AES encryption to encrypt files stored in the Firebase Database. I also integrated different libraries to enhance functionality, such as Firebase Authentication for user login, Firebase Realtime Database for storing encrypted files, and Google Play Services Location API for GPS-based access control. 🔐 Additionally, I implemented an Admin user feature using Firebase Authentication, allowing the Admin to set the time access and location access, as well as upload files to the Firebase database. Users can then log in from a normal account and access files based on the set time zone and location restrictions. 🚀 Check out the project on GitHub: GitHub -https://lnkd.in/g6N-wkrd 🛡️ It was a challenging but rewarding project that taught me a lot about data security, Java development, and integrating different libraries. Looking forward to applying these skills in future projects! #CyberSecurity #AndroidDevelopment #DataEncryption #Firebase #AccessControl #SecureVault

23

1 Comment

It was a blast working on Lagrange! Probably one of the coolest projects that I've worked on this summer. Crazy how it only took us (shoutout to Virag Tiwari and Krrish Malhotra) two weeks to recreate Cognition's Devin.ai using Google Gemini! Check out the post to see how it works #gemini #chatgpt #gpt #openai #google #devin #ai #llm #agents #microsoft #copilot #deeplearning #sde #software #machinelearning

38

10 Comments

🌟 Joining the ML-Ops Wave! 🌊 I'm excited to be attending the ML-Ops Implementation Workshop today! It's all about making machine learning models thrive in the real world. 🤖💼 Big thanks to Umer Khalid Qureshi and Tayyab Nasir for sharing their expertise. Your insights are truly inspiring and vital for our growth in the AI landscape. Let’s keep pushing the boundaries of what’s possible with AI and machine learning! 💡🔍 #MLOps #MachineLearning #DataScience #AI #Workshop #LearningIsGrowing

29