Houssein Djirdeh

Wrapped up the next piece of the puzzle, an advanced machine learning project analyzing visa application outcomes for EasyVisa! As part of the Advanced Machine Learning module in my PGP‑AIML program, I explored applicant demographics, employment details, education levels, and wage patterns to understand which factors most strongly influence visa certification decisions. From there, I built a predictive model designed to help streamline the evaluation process and support more consistent, data‑driven decision making. Project goal: Identify the key drivers of visa approval and build a model that can recommend whether an applicant’s profile is more likely to be certified or denied. What I worked on: ▪️ Exploratory data analysis to uncover patterns across applicant profiles ▪️ Data preprocessing and feature engineering ▪️ Building ensemble models using Bagging, Boosting, and Stacking ▪️ Hyperparameter tuning to improve predictive performance ▪️ Translating model insights into clear, actionable business recommendations This project strengthened my ability to work with complex classification problems and reinforced how advanced ensemble methods can improve accuracy and support fair, transparent decision processes. Skills & Tools: #EDA #DataPreProcessing #BaggingClassifier #BoostingClassifier #StackingClassifier #HyperparameterTuning #MachineLearning #BusinessInsights #PGPAIML #GreatLearning #UniversityOfTexasAtAustin https://lnkd.in/eDhiXM88

2

🚀 Excited to announce the launch of Shamsi Calendar API! 🎉 After weeks of development, I'm proud to release a comprehensive RESTful API for the Afghan Shamsi (Solar Hijri) calendar - now live at https://lnkd.in/g7TNWXys Why I built this: As Afghan developers, we've long struggled with the lack of reliable tools for our calendar system. This API changes that. 🌟 Key Features: ✅ Accurate Shamsi ↔ Gregorian date conversion ✅ Full calendar views (day, week, month, year) ✅ Complete holiday management system ✅ Trilingual support: Dari, Pashto & English ✅ Interactive Swagger documentation ✅ 151 tests with 85.6% code coverage ✅ Free and open-source 🛠️ Built with: TypeScript, Express.js, tested with Jest, and deployed with enterprise-grade security (rate limiting, CORS, Helmet) 📊 What you can do: Build Afghan calendar apps without reinventing the wheel Integrate Shamsi dates into your existing applications Access accurate holiday information Use it in any language: TypeScript, Python, Java, etc. 🎯 Perfect for: Afghan tech startups, government systems, educational platforms, mobile apps, and anyone serving the Afghan community worldwide. Try it out at the interactive documentation and let me know what you think! Contributions and feedback are welcome. #OpenSource #API #AfghanTech #ShamsiCalendar #WebDevelopment #TypeScript #SoftwareEngineering #Afghanistan

22

1 Comment

SWE Tooling Review: Lazygit (https://lnkd.in/gc2Nurt9) Git as a VCS has become the de-facto standard for engineers in 2025, and there is nothing "lazy" about it. Rebases, merging, and conflict resolution can be challenging for engineers not familiar with the intricacies of git. In response, several third-party UIs are available that flatten the learning curve for juniors by abstracting the git complexity away from users. I feel that Lazygit is a great balance between pure command line interface (how git was created and intended to be used by its creator) and some of the more "aesthetic" UIs available. Lazygit meets engineers where they are at daily: the terminal. Its TUI (Terminal User Interface) is minimal, with keyboard shortcuts for everything. It shows you the actual git commands it is running, to show how you COULD execute the commands via CLI. Where I've been using Lazygit most, is during agentic coding sessions with Opencode. Lazygit on one side, Opencode on the other. I can closely monitor what files and changes the LLM agent is making, even committing and locking in changes as it progresses.

3

Catching this talk between Lloyd Blankfein and David Solomon. Regardless of the industry, there is a lot to learn from how these leaders define "partnership."

1

From this week's episode of #Overcommitted, engineers with 2-3 years of experience hit the peak of the Dunning-Kruger Effect. Bassem Dghaidi is designing a system design course targeting exactly that demographic, helping engineers discover gaps in their knowledge and learn from others mistakes. Check out this week's episode for more insights from Bassem on technical growth and continuous learning. #SoftwareEngineering #SystemDesign #DeveloperEducation

25

3 Comments

The road to a fully autonomous SWE Agent is test-driven. Take it from someone who previously built an AI coding agent that completes frontend tasks from ticket to PR. Sohil Kshirsagar, Jun Yu Tan, Sohan Kshirsagar and I dedicated a large part of last year making sure our agent could one-shot frontend tasks for complex codebases without any human in the loop. Spoiler alert: it wasn't possible to one-shot the universe of assigned tasks entirely autonomously. Not without some form of "vibe testing" at least. Why? A big part of writing high-quality code is actually verifying that the code works as intended. This is doubly important with AI-generated code since some context only ever lives in an engineer's head. And what better way to build trust that a SWE Agent has correctly completed a task than if it wrote its own unit tests and then verified that its code passed the tests? Tusk is our answer to that. We're focused on building the best-in-class unit test generation agent. Because if you can write the best tests, you can write the highest quality code -- AI or otherwise. So yes, it turns out the fundamentals actually matter. And TDD champions everywhere have the last laugh.

34

4 Comments

A team from Tsinghua University has shattered the long-standing speed barrier in shortest path algorithms by designing a new solution that beats Dijkstra’s classical approach—without relying on sorting. Traditionally, Dijkstra’s algorithm finds shortest paths by repeatedly sorting nodes by distance, which imposes a fundamental speed bottleneck linked to sorting time. The new algorithm cleverly sidesteps this restriction by clustering boundary nodes and using selective exploration, reducing the number of candidates at each step and avoiding costly sorts. Their breakthrough also adapts Bellman-Ford techniques for directed graphs, combining randomized and deterministic innovations. This lets them solve shortest path problems faster than ever before—even for arbitrary graph weights—a feat considered impossible for 40 years. Their work won Best Paper at STOC and is set to reshape how developers approach pathfinding and graph optimization. #algorithm #programming #computerscience

4

1 Comment