frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Show HN: I indexed 37h of my videos using an RTX 4090 and local ML models in 24h

1•iliashad•1h ago
TLDR: Following my recent blog post and Hacker News post (https://news.ycombinator.com/item?id=48528029). where I ran the desktop app on my M1 Max. This time, I’m using the self-hosted version, running in Docker, with an NVIDIA RTX 4090 (24 GB of VRAM).

The content is also fundamentally more demanding: long podcast episodes with at least two faces in every frame, coding tutorials packed with on-screen text, and screen recordings. GoPro footage is mostly wide outdoor shots.

But NVIDIA was much faster than my M1 Max.

The longest video was a livestream of 3h 12m indexed in 1h 52m (4,612 frames analyzed).

You can directly see the processing jobs results in JSON format here: https://gist.github.com/IliasHad/fd64e4d331e90e57d61e95f64e8...

Comments

iliashad•1h ago
Link: https://iliashaddad.com/blog/i-indexed-37h-of-my-youtube-lib...

Show HN: I Made an SDK to work with Vapi and Retell without vendor lock in

https://voice-ai.dev
1•marchypolite•51s ago•0 comments

Knoppix

https://www.knopper.net/knoppix/index-en.html
1•hoangvmpc•1m ago•0 comments

Show HN: OpenATP: A platform for automated theorem proving in Lean

https://github.com/henryrobbins/open-atp
1•henryrobbins00•2m ago•0 comments

Pong game recompiles its own source code every frame

https://www.tomshardware.com/tech-industry/pong-game-that-recompiles-itself-every-frame-wins-the-...
1•GavinAnderegg•4m ago•0 comments

Plur1bus

https://en.wikipedia.org/wiki/Pluribus_(TV_series)
1•hosteur•4m ago•0 comments

AI agents finish a third of the job, and the math says why

https://okaneland.com/study/do-ai-agents-work-yet/
1•ermantrout•4m ago•0 comments

Show HN: PulseDB – Unified Redis and Pinecone Alternative in Python

https://github.com/gkavinrajanCodes/pulseDB
1•gkavinrajan•5m ago•0 comments

MS reveals why Windows 11 says a file is in use after closing app, plus the fix

https://www.windowslatest.com/2026/06/30/microsoft-reveals-why-windows-11-keeps-saying-a-file-is-...
2•thunderbong•5m ago•0 comments

Burn less, ship more: the case for token optimization

https://www.multiplayer.app/blog/burn-less-ship-more-the-case-for-token-optimization/
1•argoeris•6m ago•0 comments

China's humanoid robots captivated the world, a rental market shows their limits

https://www.cnn.com/2026/06/30/tech/china-humanoid-robot-ai-rental-intl-hnk-dst
1•breve•7m ago•0 comments

Show HN: NodePad – AI agent on a canvas instead of a linear chat

https://node-pad.com/
2•palazski•7m ago•0 comments

Memoirs of Extraordinary Popular Delusions and the Madness of Crowds

https://www.gutenberg.org/ebooks/24518
2•lstodd•7m ago•0 comments

Much ado about zone 2

https://pubmed.ncbi.nlm.nih.gov/40560504/
2•iamacyborg•7m ago•0 comments

Running a coding agent locally with MLX for when you hit your token cap

https://bart.degoe.de/running-a-coding-agent-locally/
1•bartdegoede•8m ago•0 comments

EYG: A Programming Language for Humans

https://crowdhailer.me/2026-06-08/a-programming-language-for-humans/
2•surprisetalk•9m ago•0 comments

Offduty – hold your Gmail and deliver it in batches on your schedule

https://offduty.me/
2•fynngentle•9m ago•1 comments

Claudeshoring

https://everythingengineer.substack.com/p/claudeshoring-enterprise-offshore-ai-projects
1•mikecarroll•10m ago•0 comments

Show HN: GSV – a personal AI computer that unifies your machines

https://github.com/deathbyknowledge/gsv
3•deathbyknowledg•11m ago•0 comments

Techno History: Archive of 475 shows with over 9000 documented tracks (German)

https://stevemasonexperience.info/
2•nor0x•13m ago•0 comments

Benchmark agent configs with a simple CLI tool

https://github.com/emiliolugo/clawmark
1•emiliolugo•13m ago•0 comments

Architecture as an executable constraint instead of documentation

https://github.com/arman-jalili/guardian-framework
3•arman-w-jalili•13m ago•0 comments

WhatsApp to let people chat with usernames

https://www.bbc.co.uk/news/articles/c1dykk3135xo
2•edward•13m ago•0 comments

Fifteen Years in Fifty Minutes

https://theshamblog.com/fifteen-years-in-fifty-minutes/
4•scottshambaugh•14m ago•0 comments

What's slowing down the AI buildout

https://www.worksinprogress.news/p/ai-is-bottlenecked-by-the-grid
2•speckx•15m ago•0 comments

Show HN: ReelsGraph – movie/TV watchlist and ML-based discovery queue, no signup

https://www.reelsgraph.com/
1•Whiskee•17m ago•2 comments

Show HN: Framein – a local work-state layer that keeps AI agents in context

https://www.framein.dev/
1•BonPPa•18m ago•0 comments

The Tech Innovations Readers Want to See in the Next 20 Years

https://www.wsj.com/tech/tech-innovations-readers-e599647b
1•thm•18m ago•0 comments

We Should Start Simulating Brains

https://www.youtube.com/watch?v=FHQfmJEpRmU
1•turizzmo•19m ago•0 comments

x86 AI Compute Extensions (Ace) Specification [pdf]

https://x86ecosystem.org/wp-content/uploads/2026/06/ACE_v1_Specification_public_1_15.pdf
1•ksec•19m ago•0 comments

Product Design Psychology

https://productdesignpsychology.com
1•thm•19m ago•0 comments