Dockerfile Basics: FROM, RUN, and CMD

I kept writing Django APIs… but something felt off. The code was working. Responses were coming. But if someone asked me: 👉 “What actually happens when a request hits your API?” …I didn’t have a clear answer. That bothered me. So I went back to basics. Not tutorials. Not copying code. Just understanding one simple flow: User → request → view → model → database → response And suddenly, things started clicking: Patient.objects.all() is not just a line of code… it’s a query hitting the database and returning structured data. request is not just a parameter… it’s literally everything the user is sending to your backend. GET, POST, PUT, DELETE are not just methods… they define how your system behaves. The biggest realization? 👉 I was focusing on “how to write code” 👉 instead of “how things actually work” Now I approach backend differently: I don’t start with code. I start with flow. And that small shift is making a huge difference. Still learning. But now it feels real. #Django #BackendDevelopment #Python #LearningInPublic #SoftwareEngineering #BuildInPublic

Decoupling logic in Django is always an interesting architectural challenge. Recently, I’ve been relying more on Django Signals to keep my models clean and enforce a strict separation of concerns. For those who haven't dug into how they work under the hood: Django signals essentially implement the Observer design pattern. There is a central dispatcher, when a specific action occurs in the application (the sender), the dispatcher routes that event to any function "listening" for it (the receiver), allowing them to execute their own logic independently. In the snippet below, I’m using the post_save signal. Whenever a new Student instance is successfully created, this receiver catches the signal and automatically generates a CreditWallet for them. Why use a signal here instead of just overriding the save() method on the Student model? It comes down to encapsulation. Overriding save() works fine for simple apps, but as a project grows, it can lead to massive, bloated models. By using signals, the Student model remains strictly responsible for student data, while the financial/wallet logic is encapsulated in its own domain. It makes the codebase much easier to maintain, scale, and test. I’m curious to hear from other developers on here: What is the most complex, creative, or technically challenging way you have utilized Django signals in a project? I'd love to learn from your experiences! #Django #Python #SoftwareEngineering #WebDevelopment #Architecture #Coding

Most beginner backend projects die in refactoring. Here's the structure I use to prevent that. When I built my Task Manager CLI, I learned this the hard way — a monolithic file that worked until it very much didn't. After refactoring, here's the structure I now start with: Before writing a single line: → Define your data model first → Identify all operations (CRUD) you'll need → Map inputs, outputs, and error states While building: → One module per concern (routes, models, utils, exceptions) → Validate inputs at the boundary — not deep inside logic → Handle errors explicitly — no silent failures Before shipping: → Test the unhappy paths, not just the happy ones → Read your own code like a stranger would This approach reduced my debugging effort by 40% on a real project. It works at any scale — from a CLI tool to a FastAPI service. What's the first thing you do when starting a new backend project? #BackendDevelopment #Python #FastAPI #SoftwareEngineering #CodingTips

Devlog Day 30 Today I revisited a question "Two Sum II" which is a very simple problem until you know the approach. The difference between Two Sum and Two Sum II is in Two Sum II the array is sorted and you have to solve the problem with O(n) TC and O(1) SC. Using two pointer approach you can solve this question. In brute force we use nested loops which loops through entire array to check if the sum of nums[i] and nums[j] equals to target. Else continue. It takes O(n2) TC and O(1) SC as we are not using any extra space for solution. In better approach we use two pointers by taking advantage of sorted order of array. We start from first and last pointer and check if the sum of them equals to target. If the sum is less we move left pointer else right. The problem in itself is not that hard but I am started analysing the pattern of such questions like if you have sorted array and want to find any element or perform any operation it is most likely targetting binary search and two pointers. Anyways later I continued my django course and today I learnt static files handling when your code is in production and even wrote a command script. Django lets you write your own command and you can access that command using python manage.py <command_name>. It really helps when you are containerizing your code and want to automate some processes like fetching static files from third party cdn and load it in locally so you dont have to request api for every reload. #DevLog #BuildInPublic #DSA #NeetCode #Stack #LearningInPublic #IndieHacker #WebDevelopment #Django #Coding #Development #Explore

🚨 The repo is now at 117k stars and climbing. All others have been DMCA'd but it's still standing. Frequently being updated. Codex is saying good things about the architecture. Either we will have a massive security event that will be remembered for decades or Open Source Claude Code... but like actually open source. (if you let codex read it, make sure you are in a safe container. beware of prompt injection whenever letting an agent interact with anything.)

Today I finally understood Docker in a practical way . Now I understand the core idea: Docker helps us package our app + dependencies + runtime environment so it runs the same way on any machine. What clicked for me: ~ Image = blueprint ~ Container = running instance of that image ~ Dockerfile = recipe to build the image Lets understand this with a simple example I have a project built with Python, NumPy, Pandas, and other libraries. [ Without Docker ] --> My friend must install Python --> Then install all required packages --> Then match versions --> Still may face setup errors [ With Docker ] --> My friend only needs Docker --> Build image --> Run container --> App works with all dependencies already packaged inside One more thing I learned: Containers do take disk space, but they are still considered lightweight compared to full virtual machines because: They share the host OS kernel They reuse image layers efficiently Also understood why teams use separate containers for frontend, backend, and database: Docker is not just a tool. It is a reliability mindset for development and deployment. Still learning, but this was a unlock for me . #Docker #DevOps #SoftwareEngineering #LearningInPublic #BeginnerDeveloper #TechJourney

🚀 pyresilience — 7 resilience patterns in 1 decorator (~1000 downloads last month 🔥) 💡 What is resilience? Your app keeps working even when dependencies fail, slow down, or overload. No crashes. No hanging. Just smart recovery. ⚠️ Pain point: Python teams often stitch together: • Retries ("tenacity") • Circuit breakers ("pybreaker") • Timeouts ("asyncio", "signal") • Rate limiting ("limits", "slowapi") • Fallbacks (custom code) 👉 These don’t coordinate → messy + inconsistent failure handling 📊 Existing tools: • "tenacity" (retries ~263.6M downloads/month) • "pybreaker" (circuit breaker ~9.6M downloads/month) 👉 Great individually, not unified ⚡pyresilience Benchmark: 🚀 pyresilience → 0.64 μs (🔥 ~10.4x faster) 🐢 tenacity → 6.64 μs 🛠️ What pyresilience does: One decorator with: ✅ Retry ✅ Circuit Breaker ✅ Timeout ✅ Fallback ✅ Bulkhead ✅ Rate Limiter ✅ Cache ➡️ Works together, not glued ➡️ Zero dependency ➡️ Sync + Async ➡️ High performance Frameworks: 🌐 FastAPI • Flask • Django 👨💻 For all Python developers 🔗 GitHub: https://lnkd.in/d-SRygNQ 🔗 PyPI: https://lnkd.in/dRg2H4D5 🔗 Docs: https://lnkd.in/dxZ4xYkw ♻️ Resharing to support the Python community 🤝 💬 How are you handling resilience in Python today? #Python #PythonDeveloper #PythonDev #BackendDevelopment #SoftwareEngineering #DistributedSystems #Microservices #SystemDesign #ResilienceEngineering #FaultTolerance #HighAvailability #ScalableSystems #PerformanceEngineering #AsyncIO #FastAPI #Django #Flask #APIEngineering #CloudNative #DevOps #SRE #OpenSource #OpenSourceProject #PyPI #PythonLibraries #DeveloperTools #TechInnovation #BuildInPublic #Programming

I got tired of Claude (and other LLMs) confidently giving me wrong function signatures, so I built a small tool that generates markdown API references from source and keeps them in your repo. (see at the end for comparison with context7) It works for Rust crates and Python packages (via runtime introspection). The output is a two-tier markdown file: curated patterns and gotchas on top, full machine-generated API reference on the bottom. Re-running the script updates the reference while keeping your curated notes intact. There's a Claude Code plugin that makes it hands-free, just say "vendor docs for tokio" and it generates the guide. After that, Claude checks the vendored docs before reaching for the internet. No more invented method signatures. Then you could just ask it to vendor a doc for you, for example today I wanted to vendor the kube crate's doc: "vendor the doc for kube" Once done, you can ask it to compare your implementation against it, and check if there are issues: "compare our usage of kube with the vendored doc" In a few seconds it'll do the check for you :) How about context7 you may ask? the main difference is that everything is local and yours. the docs are generated from your actual lockfile version and committed to your repo. no API calls, no rate limits, no downtime, works offline, works in CI, works on a plane, etc. the other thing is you own the content. you can add patterns, gotchas, project-specific notes on top of the generated reference. when you upgrade a dependency you regenerate and git shows you exactly what changed in the API. you can ask claude to adapt the documentation by adding sections that fit your usage; you can make it learn from your usage. if you're just doing a quick lookup context7 is fine, but if you're spending weeks with a library it's nice to have the docs right there in the repo, version-locked, no external dependency. https://lnkd.in/d9U2ambf

Python Tracebacks in Claude Code? Hide the Framework Frames Created by jidonglab A Django traceback for a simple TemplateDoesNotExist error is 40+ lines. 35 of those lines are Django internals — django/template/loader.py, django/core/handlers/base.py, django/middleware/common.py. Your AI doesn't need to read Django's source to fix your missing template path. But it does, ever... link https://lnkd.in/ebnuWwJt pubDate Mon, 06 Apr 2026 03:27:55 +0000

Excited to share something I've been building: Slop Report — a GitHub Action that automatically posts a code quality summary on every Python pull request. Instead of asking reviewers to hunt down coverage gaps, dependency risks, or performance regressions, Slop Report surfaces the signal directly in the PR comment thread:   Change Risk — % of modified lines covered by your test suite   Blast Radius — how many modules are affected by the change   Performance — per-test timing vs. the base branch   Maintainability — MI regression on modified files + quality of newly added code It plugs into your existing CI in minutes, never blocks a merge, and gives reviewers the data they need without leaving GitHub. Now available on the GitHub Marketplace: https://lnkd.in/eyFJvnRu Would love feedback from anyone working on Python projects or developer tooling — what metrics would you want to see next? #DevTools #GitHub #GitHubActions #Python #CodeQuality #OpenSource