3 Comments
User's avatar
L. Elaine Dazzio's avatar

For goodness sakes learn good system design principles and practices to understand what a well engineered system should look and behave like.

Expand full comment
Dyanne Gavin's avatar

I am not a software engineer, however, I love building things.

I am also a disabled senior, so my mobility is affected.

Expand full comment
LeoTalk's avatar

We talked about this issue yesterday. I summarized it in Chinese, so I've translated it into the English version here:

## TL;DR:

> In the age of AI, junior engineers can skip traditional steps and gain early access to architectural thinking and complex systems. But true growth still demands discipline, ownership, and relentless practice.

## A Junior Software Engineer’s Growth Guide in the Age of AI

> A curated reference guide inspired by @leoifuryst’s original curiosity and @dotey’s quote tweet:

> “What’s the best growth path for a Junior software engineer in the age of AI?”

>

> This guide consolidates the answers from all contributors, preserving individual insights and @mentions. Every point comes from real experiences — rigorously merged, no fabrication.

## Foundation: Build Your Knowledge Graph with AI

• AI can’t replace systematic learning

Even the best AI can’t replace broad, foundational knowledge. As a junior, you must know what exists before AI can help you go deep.

— @Jason_Young1231

• Start with a knowledge map, then break it down

Use a top-tier LLM to perform deep research, extract a competency graph, break skills into small tasks, and complete each loop fully.

— @frxiaobei

• Lower barrier for fundamentals, but higher value

Core CS knowledge like OS, networks, and compilers may be easier to access, but mastering them helps you make informed decisions later.

— @Manjusaka_Lee

## Real-World Coding: AI Pair Programming Is the New Default

• AI is the ultimate mentor

Unlike human mentors who may dodge or forget questions, AI is always available and helpful — especially for juniors.

— @dongxi_nlp

• Don’t just vibe-code — pair with AI intentionally

Let AI design your plan and explain your approach before coding. Never write code you don’t understand. Take responsibility.

— @Jason_Young1231

• Learn by building projects end-to-end

Use AI to help you scope, design, build, and deploy real apps. Go beyond tutorials — build with purpose.

— @CicidaMay, @GoshPavi

• Be a generalist first, specialist later

Role boundaries (frontend/backend/etc.) are fading. Be an engineer first. Innovation is easier now — but requires precision and decision-making.

— @Manjusaka_Lee

## Learning Methodology: Growth Is Strategic, Not Just Tactical

• Avoid the illusion of “already knowing”

With fast feedback loops, it’s easy to think you’ve learned something. True growth still demands discipline and restraint.

— @frxiaobei

• Don’t vibe aimlessly — always have intent

Copy-pasting is not engineering. The key is knowing what you’re doing and being able to continuously refine your system.

— @GoshPavi

• Many paths, one goal: open your perspective

Whether you’re grinding Leetcode, finding mentors, or learning solo — what matters is understanding the right direction.

— @arkuy99

• It’s not about your title — it’s about solving real problems

Junior or senior doesn’t matter if you can’t deliver. AI is the strongest bullshit detector.

— @mike_chong_zh

## Native AI Generation: Unique Opportunities for Young Engineers

• No legacy mindset, fast ramp-up

Young people approach AI without preconceptions. They adopt new paradigms faster and hit creative breakthroughs quicker.

— @YanyuRensheng

• From Next.js apps to LLM architectures — fast exposure

AI gives juniors early exposure to architectural thinking, domain-driven design, message queues, and more — usually reserved for seniors.

— @ZhichengWang87

## Actionable Blueprint: Learn → Build → Refine

• Step 1: Learn rapidly with AI

Use Claude/Gemini/ChatGPT to map out interview topics, system design patterns, and knowledge gaps.

• Step 2: Build with real constraints

Launch side projects with real-world requirements and deploy them. It doesn’t have to be perfect — but it must be complete.

• Step 3: Continuously refactor and improve

Rebuild older projects with better practices. Incorporate architecture patterns like DDD, queues, task schedulers.

## Final Words from the Front Lines

• AI = Equalizer + Amplifier

What was once impossible for one person is now achievable. AI shortens the path — but doesn’t eliminate the climb.

— @ZhichengWang87

• The desire to build is everything

If you’re driven by the joy of building, you’ll naturally develop the curiosity and courage to debug, ship, and improve.

— @wey_gu

• You can bite off more than you can chew — just chew slowly

AI lets you approach senior-level problems earlier. It’s okay if it’s tough. Give it time — you’ll get there.

— @GoshPavi

• “Tao” over “techniques”

It’s not about tools or job titles. It’s about whether you can ship, maintain, and improve real software that matters.

— @mike_chong_zh quoting @ryolu_

Expand full comment