Jason Voorhees
Say cheese
- Joined
- May 15, 2020
- Posts
- 77,249
- Reputation
- 224,314
This is a small roadmap that I made from my own experiences and learning about this stuff. I'm not some tech guru or but I know what is not and is not in demand atleast from an US Tech perspective and this is what I'd recommend.
This is a very intense 12-month plan. For someone starting from zero. This is a full-time commitment which I assume you can since most people here are neets. Doing this while working a day job or studying in college would likely take much longer like 18-24 months.
Months 1-3
Front-End + Al Basics
HTML/CSS/JavaScript: These are like the Alphabets of the entire Dev world. I know many people will spit at them like nigga who does HTML and javascript in 2025. Lol but no. These are important. Nobody uses vanilla javascript sure but all of the frameworks are based on it. React/Next.js are based on javascript. If you go to big tech there gheu have their own frameworks which you will not understand so if you directly jump to react.js or next.js you won't understand anything and if some newer framework like nuxt.js comes in you'll be copying Stack Overflow like a bot and crying when the interviewer asks explain how event loop works. Takes like a few weeks max. Drop your ego and just do it
React/Next js: very useful and in demand frameworks. 99% of startups run on one of these two. They are also Al-friendly frameworks and companies love them for ChatGPT plugins. It will be easy to transition to react once you master javascript first.
Also learn TypeScript while you are at it. A standard in professional settings for improving code quality and catching errors.
nextjs.org
react.dev
Al Tooling: In the big 2025 nobody writes code manually. So use Gemini and claude to code faster and generate boilerplate code and work on the logics yourself but to understand the output of those AIs you still need to learn the frameworks to correct mistakes and debug them. I suggest learn prompt engineering basics. Many free courses available. They teach you exactly how to extract and get what you want to LLMs without wasting your time
Deployment: Build small dumb projects todo app with Al twist, Al image filter, basic chatbot and throw them on Vercel or Netlify. Add OpenAl/Gemini API calls costs pennies and learn to generate liveinks. No live link = project might as well be a hallucination. Companies don't care about your localhost screenshots in 2025. No link = doesn't exist.
Months 4-6
Back-End + Al Integration
Python + Nodejs: Build REST APIs with fastAPI/Express. This is a must. I know again. Boring old stuff but will be very useful for you down the line. If you want something flex worthy and new Maybe you can try Nest also
nestjs.com
Python for anything Al-related because the entire Al ecosystem lives in Python.
AI databases:. Most people will tell you to learn databases like MonogoDB, MYSQL. All that is fine but in 2025. Vector databases are all the rage why because they store embeddings, semantic search, RAG pipelines all crucial for AI powered apps. In simpler words the entire database remembers itself and is intelligent. So I highly recommend you learn this. Pinecone is the easiest because it is serverless. Qdrant is great for self hosting but anything works
www.pinecone.io
qdrant.tech
DevOps: Don't have go a deep dive in this. Just a little bit of Devops is enough. Just the AWS/Azure basics if you have some time and money maybe get the AWS cloud practitioner certificate. You can get it in just 2 weeks of studying. Also learn Docker. Will be useful for you scaling infrastructure. All this might seem redundant for a dev but companies love seeing Devs that understand the entire deployment pipeline even if it is at a basic level. No harm in learning about it
Now you need Build and deploy a simple Al-powered SaaS (Al content generator, Al fitness coach, Al resume screener whatever). Use Next.js + FastAPI + Pinecone + Vercel + whatever.. This single project will carry your portfolio. Once this is live you can officially call yourself a developer. But to stand out from the crowd. The next months you need to focus on specialization and upskilling with more advanced stuff. At this point are done with school and are graduating to the real stuff
Month 7-10
By this stage, you'll already have a clear idea of what direction feels natural to you. You'll know what you enjoy, what you're good at, and where you want to go next. I personally liked devops and did a deep dive on that part but it's upto you some people get into Networks others grind DSA + OOPs for traditional SDE roles. But if your goal is to break into Al, here's the path I'd recommend
Learn machine learning. Now you move from "using Al" to understanding how it works. Learn the core ML ecosystem
scikit, matplotlib, Numpy, Tensorflow etc. All these are machine learning framework. Machine learning is the heart of AI. A deep dive here will separate you from people who only rely on APls. So start building models and running them on Google Colab and Kaggle notebooks. Brush up on linear algebra/probability via Khan Academy and Learn the important stuff like data preprocessing, SMOTE for imbalanced data, YOLO if you wanna do computer vision flex, basic transformers understanding, etc. You don't need to do PhD level stuff those ML researchers do just enough to not sound dumb in interviews.
Another thing I recommend you learn is AI Ethics. I know many niggas will be like AI ethics. Who tf cares and trust me many companies do. Bias detection, Fairness and privacy, jailbreak resistance. At least know the buzzwords and one or two real examples as these get asked in interview rounds for AI roles. Understanding all these principles makes you a more complete engineer
After this start building agents AI agents. Already made a thread on Google ADK.
looksmax.org
These help you build autonomous
Workflows agents, RAG pipelines and automated systems businesses actually pay for. This is the stage where your projects stop being toys and start becoming products that businesses will actually pay for. If you can build something that works without constant babysitting and delivers measurable ROl people will make it rain money for you.
www.langchain.com
After all this I highly recommend you do freelance. Recruiters love freelancers because they already know you can build stuff. Build automation tools, Al agents, ML models for clients and Help startups integrate OpenAl/Gemini/ Claude into their apps. Look for it on Upwork, Toptal or even X
Even if it's $300 to add Claude to my startup's chatbot take it. Accept even low pay at first because this experience will be invaluable into the future.
I also suggest getting proper AI certificates like AWS Machine Learning Specialty, Google Professional ML Engineer. These aren't easy courses and you will have to grind for them but they boost your CV and make you stand out especially as a young engineer. It's very rare to see these certificates being listed on someone's LinkedIn who is under 35 so a big plus.
Months 10-12
Now it is time to reep the benefits of your labour. Just to summarise you now have knowledge and expertise on
Frontend
Backend Al APIs Vector DB ML knowledge Deployed SaaS Freelance experience Live links Maybe certs
You are now fully equipped with everything to disrupt the job market and stand out from the casuals. You are effectively in the top 10-15% of all the software engineer freshers. You are the jack of all trades with just enough depth to actually build real products. You may not be a master of every domain but understand just enough to navigate anything in the software world. Now you have 2 options
Option A - Build a REAL SaaS and make money while you sleep. Like why not? You already have all the knowledge and skills to build a SaaS so why not do it. Now some niggers will ask why don't you have a SaaS buisness Mr Jason Voorhees and the answer is I don't want to. After years of grinding I want something that is a bit chill and enjoyable. I'm happy with a good stable high paying job. There was no one to tell me these things I had to figure it all myself and made a ton of mistakes also. There was no ai when I first started either. I learnt evrything from trial and error but if you do have that fire within you by all means do it. You are competent and fully equipped with all the tools to Launch a real SaaS buisness. The world is your Oyster..
Option B - Get a fat job
If you want to get a Job. Have a Portfolio with 3 live Al projects. The SaaS from month 6 is enough and GitHub with clean READMEs and live demos. Grind DSA and OOPs which are CS fundamentals. Plenty of stuff online and critical for passing technical interviews especially if you are aiming for large tech companies. You should also Target Al startups imo. VCs go feral whenever there's an AI startup so they receive insane funding which means very high salaries for you.
Another thing that is underrated is Networking. Ffs don't be a basement dweller. You didn't grind this much for being a door knob. Post your projects on Twitter & LinkedIn. Join Kaggle, HuggingFace spaces, Discord Al communities. Comment on big accounts, share your live links.
Finding a job is difficult. You'll have to face a ton of rejection, hundreds of rejection emails but I promise you that you will get a job eventually just hang in there long enough. Persevere and never stop learning. And you will get an amazing job.
This is a very intense 12-month plan. For someone starting from zero. This is a full-time commitment which I assume you can since most people here are neets. Doing this while working a day job or studying in college would likely take much longer like 18-24 months.
Months 1-3
Front-End + Al Basics
HTML/CSS/JavaScript: These are like the Alphabets of the entire Dev world. I know many people will spit at them like nigga who does HTML and javascript in 2025. Lol but no. These are important. Nobody uses vanilla javascript sure but all of the frameworks are based on it. React/Next.js are based on javascript. If you go to big tech there gheu have their own frameworks which you will not understand so if you directly jump to react.js or next.js you won't understand anything and if some newer framework like nuxt.js comes in you'll be copying Stack Overflow like a bot and crying when the interviewer asks explain how event loop works. Takes like a few weeks max. Drop your ego and just do it
React/Next js: very useful and in demand frameworks. 99% of startups run on one of these two. They are also Al-friendly frameworks and companies love them for ChatGPT plugins. It will be easy to transition to react once you master javascript first.
Also learn TypeScript while you are at it. A standard in professional settings for improving code quality and catching errors.
Next.js by Vercel - The React Framework
Next.js by Vercel is the full-stack React framework for the web.
React
React is the library for web and native user interfaces. Build user interfaces out of individual pieces called components written in JavaScript. React is designed to let you seamlessly combine components written by independent people, teams, and organizations.
react.dev
Al Tooling: In the big 2025 nobody writes code manually. So use Gemini and claude to code faster and generate boilerplate code and work on the logics yourself but to understand the output of those AIs you still need to learn the frameworks to correct mistakes and debug them. I suggest learn prompt engineering basics. Many free courses available. They teach you exactly how to extract and get what you want to LLMs without wasting your time
Deployment: Build small dumb projects todo app with Al twist, Al image filter, basic chatbot and throw them on Vercel or Netlify. Add OpenAl/Gemini API calls costs pennies and learn to generate liveinks. No live link = project might as well be a hallucination. Companies don't care about your localhost screenshots in 2025. No link = doesn't exist.
Months 4-6
Back-End + Al Integration
Python + Nodejs: Build REST APIs with fastAPI/Express. This is a must. I know again. Boring old stuff but will be very useful for you down the line. If you want something flex worthy and new Maybe you can try Nest also
NestJS - A progressive Node.js framework
NestJS is a framework for building efficient, scalable Node.js web applications. It uses modern JavaScript, is built with TypeScript and combines elements of OOP (Object Oriented Programming), FP (Functional Programming), and FRP (Functional Reactive Programming).
nestjs.com
Python for anything Al-related because the entire Al ecosystem lives in Python.
AI databases:. Most people will tell you to learn databases like MonogoDB, MYSQL. All that is fine but in 2025. Vector databases are all the rage why because they store embeddings, semantic search, RAG pipelines all crucial for AI powered apps. In simpler words the entire database remembers itself and is intelligent. So I highly recommend you learn this. Pinecone is the easiest because it is serverless. Qdrant is great for self hosting but anything works
The vector database to build knowledgeable AI | Pinecone
Search through billions of items for similar matches to any object, in milliseconds. It’s the next generation of search, an API call away.
Qdrant - Vector Database
Qdrant is an Open-Source Vector Database and Vector Search Engine written in Rust. It provides fast and scalable vector similarity search service with convenient API.
DevOps: Don't have go a deep dive in this. Just a little bit of Devops is enough. Just the AWS/Azure basics if you have some time and money maybe get the AWS cloud practitioner certificate. You can get it in just 2 weeks of studying. Also learn Docker. Will be useful for you scaling infrastructure. All this might seem redundant for a dev but companies love seeing Devs that understand the entire deployment pipeline even if it is at a basic level. No harm in learning about it
Now you need Build and deploy a simple Al-powered SaaS (Al content generator, Al fitness coach, Al resume screener whatever). Use Next.js + FastAPI + Pinecone + Vercel + whatever.. This single project will carry your portfolio. Once this is live you can officially call yourself a developer. But to stand out from the crowd. The next months you need to focus on specialization and upskilling with more advanced stuff. At this point are done with school and are graduating to the real stuff
Month 7-10
By this stage, you'll already have a clear idea of what direction feels natural to you. You'll know what you enjoy, what you're good at, and where you want to go next. I personally liked devops and did a deep dive on that part but it's upto you some people get into Networks others grind DSA + OOPs for traditional SDE roles. But if your goal is to break into Al, here's the path I'd recommend
Learn machine learning. Now you move from "using Al" to understanding how it works. Learn the core ML ecosystem
scikit, matplotlib, Numpy, Tensorflow etc. All these are machine learning framework. Machine learning is the heart of AI. A deep dive here will separate you from people who only rely on APls. So start building models and running them on Google Colab and Kaggle notebooks. Brush up on linear algebra/probability via Khan Academy and Learn the important stuff like data preprocessing, SMOTE for imbalanced data, YOLO if you wanna do computer vision flex, basic transformers understanding, etc. You don't need to do PhD level stuff those ML researchers do just enough to not sound dumb in interviews.
Another thing I recommend you learn is AI Ethics. I know many niggas will be like AI ethics. Who tf cares and trust me many companies do. Bias detection, Fairness and privacy, jailbreak resistance. At least know the buzzwords and one or two real examples as these get asked in interview rounds for AI roles. Understanding all these principles makes you a more complete engineer
After this start building agents AI agents. Already made a thread on Google ADK.
How to use Google skills to become an agent developer
You'll be hearing about the launch of Gemini 3 everywhere but one thing that went under the radar is Google skills. https://www.skills.google/ It's a full platform with 3000+ AI courses from GCP and deep mind And 150+ companies are actively hiring from it rn. Deloitte, Accenture...
looksmax.org
These help you build autonomous
Workflows agents, RAG pipelines and automated systems businesses actually pay for. This is the stage where your projects stop being toys and start becoming products that businesses will actually pay for. If you can build something that works without constant babysitting and delivers measurable ROl people will make it rain money for you.
LangChain
LangChain provides the engineering platform and open source frameworks developers use to build, test, and deploy reliable AI agents.
www.langchain.com
After all this I highly recommend you do freelance. Recruiters love freelancers because they already know you can build stuff. Build automation tools, Al agents, ML models for clients and Help startups integrate OpenAl/Gemini/ Claude into their apps. Look for it on Upwork, Toptal or even X
Even if it's $300 to add Claude to my startup's chatbot take it. Accept even low pay at first because this experience will be invaluable into the future.
I also suggest getting proper AI certificates like AWS Machine Learning Specialty, Google Professional ML Engineer. These aren't easy courses and you will have to grind for them but they boost your CV and make you stand out especially as a young engineer. It's very rare to see these certificates being listed on someone's LinkedIn who is under 35 so a big plus.
Months 10-12
Now it is time to reep the benefits of your labour. Just to summarise you now have knowledge and expertise on
Frontend
Backend Al APIs Vector DB ML knowledge Deployed SaaS Freelance experience Live links Maybe certsYou are now fully equipped with everything to disrupt the job market and stand out from the casuals. You are effectively in the top 10-15% of all the software engineer freshers. You are the jack of all trades with just enough depth to actually build real products. You may not be a master of every domain but understand just enough to navigate anything in the software world. Now you have 2 options
Option A - Build a REAL SaaS and make money while you sleep. Like why not? You already have all the knowledge and skills to build a SaaS so why not do it. Now some niggers will ask why don't you have a SaaS buisness Mr Jason Voorhees and the answer is I don't want to. After years of grinding I want something that is a bit chill and enjoyable. I'm happy with a good stable high paying job. There was no one to tell me these things I had to figure it all myself and made a ton of mistakes also. There was no ai when I first started either. I learnt evrything from trial and error but if you do have that fire within you by all means do it. You are competent and fully equipped with all the tools to Launch a real SaaS buisness. The world is your Oyster..
Option B - Get a fat job
If you want to get a Job. Have a Portfolio with 3 live Al projects. The SaaS from month 6 is enough and GitHub with clean READMEs and live demos. Grind DSA and OOPs which are CS fundamentals. Plenty of stuff online and critical for passing technical interviews especially if you are aiming for large tech companies. You should also Target Al startups imo. VCs go feral whenever there's an AI startup so they receive insane funding which means very high salaries for you.
Another thing that is underrated is Networking. Ffs don't be a basement dweller. You didn't grind this much for being a door knob. Post your projects on Twitter & LinkedIn. Join Kaggle, HuggingFace spaces, Discord Al communities. Comment on big accounts, share your live links.
Finding a job is difficult. You'll have to face a ton of rejection, hundreds of rejection emails but I promise you that you will get a job eventually just hang in there long enough. Persevere and never stop learning. And you will get an amazing job.
