frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: LayerClaw – Observability tool for PyTorch training

2•prabhavsanga•1h ago
Hi HN! I built LayerClaw (https://github.com/layerclaw/layerclaw), a local-first observability tool for PyTorch training. The problem: When training neural networks, things go wrong silently. Your loss explodes at step 47,392. Your gradients vanish in layer 12. Your GPU memory spikes randomly. By the time you notice, you've wasted hours or days of compute.

I got tired of adding print statements, manually checking TensorBoard files, and tracking down training issues after the fact. Existing tools either require cloud accounts (W&B, Neptune) or are too heavyweight for quick experiments (MLflow, TensorBoard for gradient analysis).

What LayerClaw does:

- Automatically tracks gradients, metrics, and system resources during training - Stores everything locally (SQLite + Parquet, no cloud required) - Detects anomalies: gradient explosions, NaN/Inf values, loss spikes - Provides a CLI to compare runs: `tracer compare run1 run2 --metric loss` - Minimal overhead with async writes (~2-3%)

Quick example:

```python import tracer import torch

# Initialize (one line) tracer.init(project="my-project", track_gradients=True)

# Your normal training loop model = YourModel() tracer._state.tracer.attach_hooks(model)

for batch in dataloader: loss = train_step(model, batch) tracer.log({"loss": loss.item()}) tracer.step()

tracer.finish() ```

Then analyze: `tracer anomalies my-run --auto`

What makes it different:

1. Local-first: No sign-ups, no data leaving your machine, no vendor lock-in 2. Designed for debugging: Deep gradient tracking and anomaly detection built-in (not an afterthought) 3. Lightweight: Add 2 lines to your training loop, minimal overhead 4. Works with everything: Vanilla PyTorch, HuggingFace Transformers, PyTorch Lightning

Current limitations (v0.1.0):

- CLI-only (web UI planned for v0.2) - Single-machine training (distributed support coming) - Early stage - would love feedback on what's most useful

Available now: - GitHub: https://github.com/layerclaw/layerclaw

I'm looking for contributors! I've created several "good first issues" for anyone interested in contributing. Areas where I need help: - Web UI for visualizations - Distributed training support - More framework integrations - Real-time monitoring dashboard

If you've struggled with ML training issues before, I'd love your input on what would be most valuable. PRs welcome, or just star the repo if you find it interesting!

What features would make this indispensable for your workflow?

AI that prepares you for meetings automatically

https://forms.zams.com/t/gN4eXz3mMgus
1•nddave•1m ago•1 comments

Show HN: Fundamental financial data directly in Google Sheets

https://workspace.google.com/marketplace/app/finqual/218031317945
1•myztika•2m ago•0 comments

2100 – Beyond the Horizon – A Utopian AI Short Film by the Flo Factory [video]

https://www.youtube.com/watch?v=EJNhFARaxKA
1•pdfernhout•2m ago•1 comments

Always Be Journaling (2018)

https://letterstoanewdeveloper.com/2018/12/14/always-be-journaling/
1•mooreds•3m ago•0 comments

Confidential computing and trusted execution within the agentic ecosystem

https://www.youtube.com/playlist?list=PLYGWJjYNEIt3zCGboiDjsNwHn3t2BTLAJ
1•adfm•4m ago•0 comments

Fyxer.ai – AI Email Companion

https://www.fyxer.com/
1•joeyswole•6m ago•0 comments

The network architecture of general intelligence in the human connectome

https://www.nature.com/articles/s41467-026-68698-5
2•bookofjoe•6m ago•0 comments

Show HN: Inklings – Handwritten family notes turned into a printed book monthly

https://inklings.social
1•archaeal•9m ago•1 comments

Mistral Is Not a European Alternative (Yet) – Here's Why

https://www.xprivo.com/blog/en/mistral-is-not-a-european-alternative/
1•petethomas•10m ago•0 comments

WorkHourly tracking app for people who work with their hands

1•bukhtin•14m ago•0 comments

Severo Ornstein

https://en.wikipedia.org/wiki/Severo_Ornstein
2•exizt88•17m ago•0 comments

Show HN: I launched a MIDI controller and IG banned me 48 hours later

3•harmonyresearch•19m ago•0 comments

Show HN: Lost in a Wikipedia rabbit hole? Pop out and survey the field

https://chromewebstore.google.com/detail/constellations/nphipbpoephgjgapmeanccnaikljggkg
1•johndimm•22m ago•0 comments

A Tale of AI and Apples

https://arbinquiry.com/a-tale-of-ai-and-apples/
1•ateesdalejr•23m ago•1 comments

'jdupes', or how I unexpectedly became a better programmer (2015)

https://www.jodybruchon.com/2015/08/08/finding-duplicates-faster-the-story-of-jdupes-or-how-i-une...
2•indigodaddy•25m ago•0 comments

Order Imposing and Reporting Requirements on Certain Finance Institutions in MN

https://www.federalregister.gov/documents/2026/01/13/2026-00449/geographic-targeting-order-imposi...
1•duxup•27m ago•1 comments

The Pinhole view of the value of AI as simply reducing payroll costs is wrong

https://tidyfirst.substack.com/p/the-pinhole-view-of-ai-value
1•KentBeck•29m ago•0 comments

Indiewebify.me? Yes Please

https://blog.rickardlindberg.me/2026/02/04/indie-webify-me-yes-please.html
1•todsacerdoti•30m ago•0 comments

Ask HN: Have you started to use "->" when you type after working with LLMs?

2•FailMore•31m ago•0 comments

The Boomcession: Why Americans Hate What Looks Like an Economic Boom

https://www.thebignewsletter.com/p/the-boomcession-why-everyone-but
3•connor11528•33m ago•0 comments

The Trigger in the Haystack: Extracting and Reconstructing LLM Backdoor Triggers

https://arxiv.org/abs/2602.03085
1•maltalex•34m ago•0 comments

Write for Yourself, and Wisdom Will Follow

https://moretothat.com/writing-and-wisdom/
1•herbertl•35m ago•0 comments

Staring into the abyss as a core life skill (2022)

https://www.benkuhn.net/abyss/
2•jger15•35m ago•0 comments

Why More Companies Are Recognizing the Benefits of Keeping Older Employees

https://longevity.stanford.edu/why-more-companies-are-recognizing-the-benefits-of-keeping-older-e...
4•andsoitis•36m ago•0 comments

We Used To Build Things. What Happened?

https://garryslist.org/posts/we-used-to-build-things-what-happened
3•rmason•37m ago•0 comments

Accountability, APIs, and the United States of Surveillance

https://netapinotes.com/accountability-apis-and-the-united-states-of-surveillance/
1•johnathandos•39m ago•0 comments

Show HN: Quibble – Adversarial AI document review using Codex and Claude

https://github.com/mfelix/quibble
1•threekindwords•41m ago•0 comments

The Inevitable Future of Stablecoins

https://layerzero.network/blog/the-inevitable-future-of-stablecoins
2•lawrenceyan•46m ago•0 comments

NDP wants Carney to kill U.S. fighter jet contract in favour of Swedish aircraft

https://www.ctvnews.ca/politics/article/ndp-wants-carney-to-kill-us-fighter-jet-contract-in-favou...
9•c420•47m ago•2 comments

AI startup Adaption Labs raised $50M

https://fortune.com/2026/02/04/adaption-labs-50-million-seed-funding-emergence-captial-sara-hooke...
2•sudohalt•50m ago•0 comments