IdeaWell

Project ideas from Hacker News discussions.

Micron CEO Says Memory Supply Will Be Much Tighter in 2027 and 2028 Than in 2026

Original Article

Hacker News Discussion

📝 Discussion Summary (Click to expand)

Four prevalent themes in the discussion

  1. AI‑driven RAM shortage and delayed supply expansion
    The surge in AI workloads has pushed memory demand far above current supply, and new fabs won’t come online until 2027‑2028.

    “Demand is way up, supply cannot keep up, it will take some time to increase supply. Micron is working on increasing that supply… It should start production in 2027… a second massive fab… scheduled for 2028.” – [tzs]
    “One factor constraining supply for RAM in the short‑to‑medium term is that memory manufacturers is that DDR5 (and assumedly DDR6) capacity is being moved to the much more profitable HBM.” – [jmyeet]

  2. Consumer purchasing behavior: buy now vs. wait
    Many commenters discuss whether to purchase memory immediately amid rising prices or hold off for a potential future dip.

    “Yeah he doesn't want people delaying purchasing decisions.” – [Neil44]
    “I've been wanting to upgrade my workstation but the DDR5 RDIMM sticks are unbelievably expensive… now I think I'll continue wait at least for the beginning of next year.” – [embedding‑shape]

  3. Speculative AI bubble and its possible burst
    A large share of the current memory demand is seen as built on speculative AI investments; participants warn of a bubble that could collapse and reshape the market.

    “Anthropic is expected to spend 500B of money it doesnt have… a lot of demand for memory is built on a speculative foundation.” – [bwfan123]
    “When the AI bubble bursts, Nvidia is going to be Enron.” – [Covzire]
    “I think supply will be tight until at least the AI market crashes, which could take years or happen within months.” – [jmyeet]

  4. Software optimization as a silver lining
    High RAM costs are viewed as an incentive to write leaner, less bloated software, reducing reliance on ever‑larger hardware.

    “If there’s a silver lining, I hope it leads to more optimisation, less Electron, and less bloat.” – [dannyw]
    “That's the one silver lining… I've heard one too many times 'just buy more RAM' as a defense for resource intensive graphical programs using Electron or modern browsers.” – [ocd]

🚀 Project Ideas

TabLiberator

Summary

  • An advanced browser extension that aggressively suspends inactive tabs, saves their state to disk, and restores them instantly, targeting the runaway memory usage of modern browsers.
  • Core value proposition: dramatically reduce Firefox/Chrome RAM footprint while preserving seamless tab recovery, addressing users who hit 10+ GB usage after hours of browsing.

Details

Key Value
Target Audience Power users, developers, and anyone with limited RAM who keeps many tabs open
Core Feature Intelligent tab suspension with predictive pre‑unload, state compression, and one‑click restore
Tech Stack WebExtensions API, IndexedDB for state storage, Rust/WASM for compression heuristics
Difficulty Medium
Monetization Hobby

Notes

  • HN users praised tab suspension: "The New Tab Suspender plugin has my 50+ Firefox tabs at 2.2GB of ram usage right now." – undersuit
  • Provides a practical utility for discussion on memory‑bloat solutions and could inspire browser vendors to adopt similar limits natively.

MemLimit GUI

Summary

  • A cross‑platform desktop application that lets users set per‑process memory limits via cgroups (Linux) or job objects (Windows/macOS) through a simple graphical interface, with real‑time usage feedback and alerts.
  • Core value proposition: give non‑technical users the power to constrain memory‑hungry apps (e.g., Firefox, Electron) without wrestling with command‑line tools.

Details

Key Value
Target Audience Linux desktop users, Windows/macOS power users, and anyone frustrated by unchecked app memory growth
Core Feature Drag‑and‑drop process selection, configurable hard/soft limits, notifications when limits are approached
Tech Stack Electron (or Tauri) for UI, libcontainer/cgroups v2, native bindings for Windows Job Objects
Difficulty Medium
Monetization Hobby

Notes

  • HN comment highlighted a manual workaround: "On linux do something like systemd-run --user --scope -p MemoryHigh=XG -p MemoryMax=X+2G firefox" – StillBored
  • Enables broader discussion about OS‑level memory governance and could be adapted into distro‑specific utilities.

TinyLLM Runtime

Summary

  • An open‑source inference engine that aggressively quantizes (4‑bit or lower) and prunes LLMs to run on devices with ≤1 GB RAM, providing a simple HTTP/API server for local chat or code completion.
  • Core value proposition: bring the ability to run useful language models on aging laptops, Raspberry Pi, or even SBCs, satisfying the nostalgia for hackable, low‑resource computing.

Details

Key Value
Target Audience Developers, hobbyists, and educators who want to experiment with LLMs on low‑end hardware
Core Feature Automatic model quantization, KV‑cache caching to disk, and batch‑size‑1 inference optimized for low RAM
Tech Stack C++ core with bindings, GGUF format support, optional Vulkan compute for acceleration
Difficulty High
Monetization Revenue‑ready: API‑service tiers (free hobby tier, paid for higher throughput)

Notes

  • Users expressed desire to run LLMs on tiny machines: "I do not, no; can you give me some context for the question? :)" and jokes about running an LLM on a C64 – formvoltron & NikolaNovak
  • Sparks conversation about democratizing AI access and mirrors the HN theme of reclaiming control over personal compute.

MemProfiler Lite

Summary

  • A lightweight, sampling‑based memory profiler for native applications (C/C++/Rust) that visualizes allocation hotspots, thread counts, and retention times via a web UI, designed to run with minimal overhead.
  • Core value proposition: help developers identify why their software consumes excessive RAM or spawns many threads, directly addressing Firefox‑style memory growth concerns.

Details

Key Value
Target Audience System programmers, performance engineers, and open‑source maintainers
Core Feature Low‑overhead sampling, flame‑graph of allocations, thread‑stack explorer, export to JSON
Tech Stack Rust agent using perfetto/eBPF, React + TypeScript frontend
Difficulty Medium
Monetization Hobby

Notes

  • HN user noted thread proliferation: "I can see in ProcessExplorer how many threads it starts & keeps, and for whatever reason this is always a lot" – KellyCriterion
  • Provides tangible utility for discussions on GC vs manual memory management and could be integrated into CI pipelines to catch regressions.

DeviceHacker Kit

Summary

  • A curated set of open‑source tools and guides to unlock bootloaders, flash custom firmware, and gain root access on common consumer devices (smart TVs, routers, IoT appliances), enabling users to replace bloatware with lean, memory‑efficient alternatives.
  • Core value proposition: restore the ability to truly own, understand, and hack one's hardware, countering the trend of locked‑down consumer electronics.

Details

Key Value
Target Audience Tinkerers, privacy advocates, and developers seeking to repurpose or optimize embedded devices
Core Feature Device‑specific exploit scripts, custom firmware images (e.g., OpenWrt, LibreELEC), and a web‑based dashboard for managing flashed devices
Tech Stack Python/ Bash for scripts, Yocto/Project‑based build system, Vue.js dashboard
Difficulty High
Monetization Hobby

Notes

  • HN lamented loss of device ownership: "plus the notion that you can truly own, understand, and hack your device, which sadly is gone from vast majority of tablets and phones and tvs and computers and even washing machines." – NikolaNovak
  • Encourages practical hacking projects and discussion on right‑to‑repair, aligning with the community's interest in low‑level control and optimization.

Read Later