robmay

5 ms·

1.
Is Compounding Inference as Powerful as Compounding Interest?(blog.neurometric.ai)
1 points by robmay 3d ago|0 comments
2.
Show HN: A SLM Optimized for Tool Calling(blog.neurometric.ai)
2 points by robmay 20d ago|0 comments
3.
Gemma 4B vs. Gemini Flash on Tool Calling(neurometric.substack.com)
3 points by robmay 1mo ago|0 comments
4.
Half Our SLM Benchmark 'Failures' Contained the Right Answer(neurometric.substack.com)
2 points by robmay 1mo ago|0 comments
5.
A Systems Design View of Why Voter ID Is Irrelevant(sharpermind.substack.com)
2 points by robmay 2mo ago|1 comments
6.
GonzoCapital – The Inside VC Stories(gonzocapital.net)
7 points by robmay 2mo ago|1 comments
7.
Pros and Cons of the Harbor Framework(neurometric.substack.com)
3 points by robmay 3mo ago|0 comments
8.
by robmay 3mo ago
Definitely - when you consider how varied inference workloads will be, and the different ways to minimize costs - better prompting, SLMs, different chips, batching, etc, there will be tons of opportunity
9.
by robmay 3mo ago
Best HN comment in a long time.
10.
Tokenminning: Because Tokenmaxxing Is a Bad Idea(tokenminning.com)
4 points by robmay 3mo ago|4 comments
11.
Lean Inference: Lean Manufacturing Principles Applied to AI(neurometric.substack.com)
5 points by robmay 4mo ago|0 comments
12.
A Marketplace of Fine Tuned SLMs for Agentic Tasks(marketplace.neurometric.ai)
3 points by robmay 4mo ago|1 comments
13.
by robmay 5mo ago
Most people don't have the team and time to do heavy token efficiency engineering. But that's all we do. marketplace.neurometric.ai has a bunch of task specific small models, and we charge flat monthly fees. We bear the token ri
14.
HarborMaster: Run Harbor at Scale(neurometric.substack.com)
2 points by robmay 5mo ago|0 comments
15.
HX Is the New UX: Designing for the Harness Experience(neurometric.substack.com)
1 points by robmay 5mo ago|0 comments
16.
Yo-GPT: A Model That Can Say "Yo"(neurometric.ai)
3 points by robmay 6mo ago|0 comments
17.
We Built 115 Free SLMs for Specific Agentic Tasks(neurometric.substack.com)
3 points by robmay 6mo ago|0 comments
18.
Show HN: Small Model Marketplace W 100M Tokens Free(marketplace.neurometric.ai)
1 points by robmay 6mo ago|0 comments
19.
Contrarian AI Investment Theses(investinginai.substack.com)
2 points by robmay 6mo ago|1 comments
20.
The Research Behind Our Auto-SLM Generator(neurometric.substack.com)
1 points by robmay 6mo ago|0 comments
21.
Show HN: Clawbake: Multi-User Instance Management for OpenClaw(neurometric.substack.com)
4 points by robmay 6mo ago|0 comments
22.
Training Qwen 4B to Beat Large Models on Work Tasks(neurometric.substack.com)
16 points by robmay 7mo ago|0 comments
23.
by robmay 8mo ago
Lemurian Labs is working on this https://www.lemurianlabs.com/
24.
Research Papers on SLMs(neurometric.substack.com)
3 points by robmay 8mo ago|0 comments
25.
Research Papers Defining the SLM Revolution(neurometric.substack.com)
1 points by robmay 8mo ago|0 comments
26.
by robmay 10mo ago
Don't you think Elon Musk and his influence on Twitter counts as an elite? I'd argue the elites are the most followed people on social
27.
Dynamically Choosing a TTC algorithm has huge performance implications(neurometric.substack.com)
1 points by robmay 11mo ago|1 comments
28.
How Test Time Compute Algorithms Impact Different Models(neurometric.substack.com)
2 points by robmay 11mo ago|0 comments
29.
What I Learned Reading Abstracts of 200 Test Time Compute Research Papers(neurometric.substack.com)
1 points by robmay 1y ago|1 comments
30.
by robmay 1y ago
TL;DR - There is no one size fits all, It's a massively underexplored space, and it may end up being more important than the training side of AI.

More ›