MacBook vs Mac Mini vs Windows laptop vs custom PC for an AI Engineer — ₹1.5L budget
retained text
Hey everyone, I'm looking to buy a **personal computer** and would appreciate some advice from people who have experience with Macs, Windows/Linux machines, and software/AI development. I'm an **AI/ML engineer**, mainly working with LLM applications, NLP, information retrieval, Python, Docker, etc. I don't generally run large models locally. For heavy GPU workloads, I rent GPUs/cloud machines when needed. # My current setup My company provides me with a work laptop: * 32 GB RAM * Intel i5 * 1 TB SATA III SSD * Around 7–8 years old * Windows It's still perfectly usable, but I'm starting to hit its limits. At peak workloads, my **RAM usage reaches around 90–93%**, but the bigger issue is that my **CPU regularly reaches \~95% utilization**. My typical workload includes: * Multiple Docker containers * Multiple Chrome windows/tabs * VS Code / IDEs * Python environments * Databases * Local services * Various development tools running simultaneously So I'm looking for a personal machine that gives me considerably more headroom. # My situation I work from home but travel regularly between **two cities**. I already have external monitors, microphones, etc. in both places. I normally carry my webcam, mouse and keyboard with me. The main problem is that I always have to carry my **work laptop** with me because I need it for work. So I'm considering getting a separate personal machine. My budget is around **₹1.5 lakh**, and ideally I'd like this machine to last **7–8 years**. My current work laptop is already around that age and is still working fine, so I'm hoping for a similar lifespan from this purchase. I'm also not necessarily looking for the absolute maximum performance today. I'd rather have a machine/platform that remains useful for a long time and, where possible, can be upgraded. # Option 1: MacBook I'm considering a MacBook because: * I already use an iPhone and iPad * The Apple ecosystem would be useful * macOS gives me a Unix-based development environment * Good battery life would be useful for travelling * Apple Silicon has excellent performance/watt * I don't necessarily need a dedicated GPU locally With my budget, **48/64 GB configurations aren't realistic**, so I'm probably looking at **16 GB or 24 GB RAM**. This is where I'm particularly unsure. **For an AI/ML software engineer doing mostly Docker/Python/LLM application development, which Apple Silicon chipset would you recommend at this budget?** Would an M-series base/pro chip with 16/24 GB be enough, or is it worth spending more on a higher-tier chip while getting less RAM? I also don't expect to run large LLMs locally. For serious GPU workloads, I'd continue renting NVIDIA GPUs. # Option 2: Mac Mini Another idea is getting a **Mac Mini** and keeping it permanently at home. I could connect it to: * Ethernet * Power * My existing monitor/peripherals Then when I'm in the other city, I could simply **SSH into it** and use it as my personal development machine. This seems attractive because I don't actually need to physically carry my personal machine every day. The downside is obviously that I'm dependent on my home internet connection and remote access. I'm also wondering whether a Mac Mini makes more sense than a similarly priced MacBook if portability isn't essential. # Option 3: Windows laptop with 32 GB RAM The third option is simply buying a **Windows laptop with 32 GB RAM** and a reasonably powerful modern CPU. I could potentially get more raw hardware for the money than with a MacBook. The downside is that I'd be staying with Windows rather than getting the Unix/macOS development environment. I also don't necessarily need a dedicated NVIDIA GPU. For the heavy AI workloads where I actually need CUDA, I generally rent GPUs in the cloud rather than running models locally. # Option 4: Custom-built Windows desktop The fourth option I'm considering is building a **custom Windows desktop/workstation** instead of buying a laptop. The idea would be to spend the initial budget on: * A **good CPU** * At least **32 GB RAM**, preferably with room to upgrade to 64/128 GB later * A good motherboard/platform with upgradeability * Fast NVMe storage * A relatively **low-end GPU initially** Then I could gradually upgrade the machine over the next few years instead of replacing the entire computer. For example, I could start with a modest GPU and eventually replace it with a much more capable NVIDIA GPU when GPU workloads become more important. This is also interesting because, unlike the Mac options, I could potentially use the machine for **local LLM inference**. It would be nice if the initial/future configuration could run something like **Qwen3-8B or Qwen3-27B with 4-bit quantization** locally, although this isn't a hard requirement. The main advantage I see is that I could get a strong CPU and plenty of RAM now, and then upgrade the GPU, RAM, storage, etc. over time. The downside is obviously that it's a **desktop**, so it wouldn't be something I could physically take between my two cities. I'd probably have to use remote access when I'm away from home. So I'm also wondering whether, given my situation, **a custom desktop + remote access** makes more sense than spending the entire ₹1.5 lakh on a laptop. # What would you choose? Given this use case, which direction would you go? **A)** MacBook — 16/24 GB **B)** Mac Mini — 16/24 GB + SSH/remote access **C)** Windows laptop — 32 GB RAM **D)** Custom Windows desktop — strong CPU + 32 GB+ RAM + low-end GPU, with upgrades later **E)** Something else And specifically for the Mac options: >**Which Apple Silicon chipset would you choose for this workload and ₹1.5 lakh budget?** For the custom PC option, I'd also be interested in what kind of **CPU + motherboard/platform + RAM + entry-level NVIDIA GPU** you'd choose if the goal is to keep the machine for 7–8 years and upgrade it over time. # My priorities 1. **7–8 year useful lifespan** 2. More CPU headroom than my current laptop 3. Enough RAM for Docker + Chrome + IDEs + development tools 4. Good software/AI development experience 5. Unix-like environment would be a plus 6. Reliability 7. Upgradeability would be a major plus 8. Staying around **₹1.5 lakh** I don't expect the machine to run large LLMs locally. For serious GPU workloads, I'd continue using rented/cloud NVIDIA GPUs. That said, being able to run something like **Qwen3 locally with 4-bit quantization** would be a nice bonus, especially with the custom PC option. Would especially appreciate opinions from people who have actually used **MacBooks, Mac Minis, Windows laptops, or custom Linux/Windows workstations for software/ML development for several years.** **Edit:** Grammar fixes and writing with ChatGPT.
document fetched 2026-09-21 00:24 · canonical https://reddit.com/r/developersIndia/comments/1wlby1d/macbook_vs_mac_mini_vs_windows_laptop_vs_custom