Hi Jeeva! First, take a deep breath: having this massive of a tech stack at age 15 is incredible, and you are far ahead of most people your age.What you are experiencing with copy-pasting your own code isn't "fake coding"—it is exactly how professional developers work. In the industry, we call this creating templates, components, and boilerplate code. Reusing your own working code to scale a project is smart engineering, not a sign of weakness.To beat the imposter syndrome and transition from a framework user to a deeply skilled engineer, you need to master computer science fundamentals. Here is the exact roadmap and order you should follow, paired with fast-paced, highly engaging creators who match your learning style.

1. Computer Science & Low-Level FundamentalsBefore jumping straight into hard DSA, you need to understand how computers actually handle memory, execution, and data. Learning a lower-level language will make your JavaScript and Python code ten times better because you'll finally understand what those languages are doing under the hood.What to learn: C or C++, Pointers, Memory Management (Stack vs. Heap), and Compilation.Highly Engaging Sources:ThePrimeagen (YouTube): A Netflix engineer who is incredibly high-energy, fast-paced, and entertaining. He focuses heavily on low-level concepts, Vim, and engineering mindset.Low Level JavaScript / Tsoding Daily (YouTube): Very raw, engaging, and fast-paced coding that dives deep into how things work under the hood.Harvard’s CS50x (Free Course): Watch just the first 5 weeks (the C language portion). David J. Malan is arguably the most stimulating and theatrical computer science professor in the world

2. Data Structures and Algorithms (DSA)Once you understand pointers and memory in C/C++, JavaScript and Python DSA will feel like a breeze. DSA trains your brain to solve complex logic puzzles efficiently.What to learn: Arrays, Linked Lists, Stacks/Queues, Trees, Graphs, Sorting algorithms, and Big-O Notation (time/space complexity).Highly Engaging Sources:NeetCode (YouTube): The absolute gold standard for DSA. His explanations are visually stimulating, crystal clear, completely object-oriented, and entirely on-point with zero fluff.Clément Mihailescu (YouTube): Fast-paced, punchy content about algorithm design and coding interview mentalities.

3. Software Architecture & Design PatternsTo stop feeling like you are just hacking things together, you need to learn how to clean up your code structure so it can scale without turning into a mess.What to learn: OOP (Object-Oriented Programming) vs Functional Programming, SOLID Principles, and Common Design Patterns (Factory, Singleton, Observer).Highly Engaging Sources:Fireship (YouTube): The king of high-stimulation, hyper-edited, fast-paced programming videos. His "Code This, Not That" and "Design Patterns in 100 Seconds" series are perfect for you.ArjanCodes (YouTube): Highly structured, engaging refactoring videos that show you how to take messy code and turn it into professional architecture

4. System Design & Databases (Advanced)Since you already know Node, Express, and MongoDB, you should level up your backend game by moving away from just "making things work" to "making things scale."What to learn: Relational Databases (PostgreSQL/MySQL) vs NoSQL, Indexing, Caching (Redis), REST vs GraphQL vs gRPC, and System Architecture.Highly Engaging Sources:ByteByteGo (Alex Xu on YouTube): Incredible, fast-paced animated visual guides to complex system architecture.Hussein Nasser (YouTube): A bit more technical, but deeply passionate and engaging videos dissecting backend engineering, network protocols, and database internals. I hope my answer helps yoy this time Jeeva all the best !!!

Yo Jeeva, respect for even asking this at 15, most people your age aren't thinking this deep about it. Here's my honest take:

First, kill the imposter syndrome talk for a sec. You've been coding since 12-13 and already know a full MERN-ish stack plus you're touching ML. That's not "nothing." The reason it feels like nothing is because you're comparing yourself to some imaginary "real programmer" instead of looking at where you were a year ago. Also copy-pasting your own code and modifying it for new features? That's literally how devs work. Nobody writes everything from scratch every time. That's called reusing patterns, not "not being a real programmer."

On what to learn next:

Skip jumping into C/C++ right now just for "memory." You don't need that yet. What you actually need is DSA, and you can learn DSA in JS or Python, languages you already know. Don't add a new language on top of a new concept, that's just extra friction. Learn DSA first in a language you're comfortable in. If later you want to go low-level (systems programming, game engines, embedded stuff) then pick up C/C++. But for now it'll just slow you down.

Order I'd actually do:

1. DSA basics (arrays, strings, hashmaps, recursion, then trees/graphs later) — this is the actual "makes you a better dev" skill people mean when they say DSA. It's less about the algorithms and more about training your brain to break problems down.
2. While doing that, build ONE real project that's yours, not a tutorial clone. Doesn't have to be big. Something with an actual problem you're solving. This fixes the "I can't create anything" feeling way faster than any course will.
3. After that, go back and actually get comfortable in TS/Node/Redux again since you said that's rusty. Rebuild a small project in it, don't just "review notes."
4. THEN come back to ML more seriously if you still want it, once your fundamentals are solid.

Resources that are fast-paced, not boring:

• NeetCode (YouTube + neetcode.io) — DSA explained super clean and to the point, no fluff
• Fireship — for quick, high energy overviews of concepts/tools, great for when you don't want a 3 hour lecture
• CS50 (Harvard, free, on YouTube/edX) — genuinely entertaining and well produced if you want real CS fundamentals, not boring at all
• Theo (t3.gg) — good for web dev / TS practical stuff, fast and opinionated
• The Coding Train — great if you want a fun way to build DSA/logic intuition through visuals

Honestly bro you have 8 years like you said, you don't need to rush. You're already ahead. Just stop tutorial hopping and build stuff, even small broken stuff. That fixes the "I feel like a fraud" thing faster than any resource list will.

Hey Jeeva, you're already way ahead for 15! Since others covered DSA and roadmaps, here is one practical thing that will instantly kill the "I only copy-paste" feeling:

1. Stop starting from blank files, contribute to Open Source: Look at intermediate open-source repositories built with Next.js or React. Try reading other developers' code, understanding their PRs, and fixing a "good first issue." Reading professional code teaches structure far better than building solo clone apps.

2. Build an API-driven real-world tool: Instead of standard CRUD or wiki sites, build something that relies on real-time data or background processing (e.g., a web scraper, a rate-limited API gateway, or a bot). This forces you to think about error handling, edge cases, and asynchronous programming.

3. Write tests: If you want to feel like a real engineer, write Unit/Integration tests (using Vitest or Jest) for your existing code. Testing forces you to write cleaner, modular code and proves your logic actually works without guessing.