Neurometric AI
@neurometric.bsky.social
Programmable inference time compute for open-source LLMs. Unlock new reasoning, performance, and control. | neurometric.ai | neurometric.substack.com | info@neurometric.ai | https://inferencetimetactics.podbean.com
New episode of Inference Time Tactics: Benchmarking Generalization: How AI Learns Beyond Training Data
Rob May and Cooper from NeuroMetric talk with AI researcher Yash Sharma about how models actually generalize and why they behave unpredictably in novel scenarios.
inferencetimetactics.podbean.com
Rob May and Cooper from NeuroMetric talk with AI researcher Yash Sharma about how models actually generalize and why they behave unpredictably in novel scenarios.
inferencetimetactics.podbean.com
Inference Time Tactics | NeuroMetric AI
A podcast exploring the emerging field of inference-time compute—the next frontier in AI performance. Hosted by the Neurometric team, we unpack how models reason, make decisions, and perform at runtim...
inferencetimetactics.podbean.com
November 6, 2025 at 10:10 PM
New episode of Inference Time Tactics: Benchmarking Generalization: How AI Learns Beyond Training Data
Rob May and Cooper from NeuroMetric talk with AI researcher Yash Sharma about how models actually generalize and why they behave unpredictably in novel scenarios.
inferencetimetactics.podbean.com
Rob May and Cooper from NeuroMetric talk with AI researcher Yash Sharma about how models actually generalize and why they behave unpredictably in novel scenarios.
inferencetimetactics.podbean.com
Apple's "To Infinity and Beyond" paper shows where AI systems are headed next.
State Space Models like Mamba cannot generalize on their own. But when given tools at runtime (search, memory, code execution) they can reason and scale far beyond their training range.
State Space Models like Mamba cannot generalize on their own. But when given tools at runtime (search, memory, code execution) they can reason and scale far beyond their training range.
Apple’s “To Infinity and Beyond” Shows Where AI Systems Are Headed Next
Competition is shifting from model architecture to runtime strategy. The key question is no longer which model is smarter, but which system makes better decisions while running.
open.substack.com
October 22, 2025 at 6:26 PM
Apple's "To Infinity and Beyond" paper shows where AI systems are headed next.
State Space Models like Mamba cannot generalize on their own. But when given tools at runtime (search, memory, code execution) they can reason and scale far beyond their training range.
State Space Models like Mamba cannot generalize on their own. But when given tools at runtime (search, memory, code execution) they can reason and scale far beyond their training range.
New ITT episode: Solving the Cold Start Problem in AI Inference
Rob May, Calvin Cooper, and Byron Galbraith talk with Prashanth Velidandi (@pmv_inferx) about InferX, serverless inference, idle GPUs, and how infra innovation is defining the next era of AI.
Listen: inferencetimetactics.podbean.com
Rob May, Calvin Cooper, and Byron Galbraith talk with Prashanth Velidandi (@pmv_inferx) about InferX, serverless inference, idle GPUs, and how infra innovation is defining the next era of AI.
Listen: inferencetimetactics.podbean.com
Inference Time Tactics | NeuroMetric AI
A podcast exploring the emerging field of inference-time compute—the next frontier in AI performance. Hosted by the Neurometric team, we unpack how models reason, make decisions, and perform at runtim...
inferencetimetactics.podbean.com
October 3, 2025 at 1:05 AM
New ITT episode: Solving the Cold Start Problem in AI Inference
Rob May, Calvin Cooper, and Byron Galbraith talk with Prashanth Velidandi (@pmv_inferx) about InferX, serverless inference, idle GPUs, and how infra innovation is defining the next era of AI.
Listen: inferencetimetactics.podbean.com
Rob May, Calvin Cooper, and Byron Galbraith talk with Prashanth Velidandi (@pmv_inferx) about InferX, serverless inference, idle GPUs, and how infra innovation is defining the next era of AI.
Listen: inferencetimetactics.podbean.com
Check out Inference Time Tactics with Rob May, Calvin Cooper and special guest Pawan Deshpande - AI researcher and founder turned product leader and angel investor.
In this episode: MIT research, why inference time decisions matter, how to evaluate agents, and where durable value sits in the stack.
In this episode: MIT research, why inference time decisions matter, how to evaluate agents, and where durable value sits in the stack.
September 30, 2025 at 11:32 PM
Check out Inference Time Tactics with Rob May, Calvin Cooper and special guest Pawan Deshpande - AI researcher and founder turned product leader and angel investor.
In this episode: MIT research, why inference time decisions matter, how to evaluate agents, and where durable value sits in the stack.
In this episode: MIT research, why inference time decisions matter, how to evaluate agents, and where durable value sits in the stack.
We are excited to share the launch of ITC Studio alpha.
This tool makes it simple to test inference time compute strategies across LLMs.
Why it matters:
ITC impacts cost, performance, and accuracy
Experimentation accelerates AI innovation
Intelligence is moving to the systems level
This tool makes it simple to test inference time compute strategies across LLMs.
Why it matters:
ITC impacts cost, performance, and accuracy
Experimentation accelerates AI innovation
Intelligence is moving to the systems level
neurometric.substack.com
September 29, 2025 at 8:22 PM
We are excited to share the launch of ITC Studio alpha.
This tool makes it simple to test inference time compute strategies across LLMs.
Why it matters:
ITC impacts cost, performance, and accuracy
Experimentation accelerates AI innovation
Intelligence is moving to the systems level
This tool makes it simple to test inference time compute strategies across LLMs.
Why it matters:
ITC impacts cost, performance, and accuracy
Experimentation accelerates AI innovation
Intelligence is moving to the systems level
New Inference Time Tactics episode: Drag, Drop, and Deploy
Rob May, Calvin Cooper, Byron Galbraith, and Dave Rauchwerk share updates on the NeuroMetric Inference Time Compute Studio and why AI is moving from scaling single models to full AI system orchestration.
inferencetimetactics.podbean.com
Rob May, Calvin Cooper, Byron Galbraith, and Dave Rauchwerk share updates on the NeuroMetric Inference Time Compute Studio and why AI is moving from scaling single models to full AI system orchestration.
inferencetimetactics.podbean.com
Inference Time Tactics | NeuroMetric AI
A podcast exploring the emerging field of inference-time compute—the next frontier in AI performance. Hosted by the Neurometric team, we unpack how models reason, make decisions, and perform at runtim...
inferencetimetactics.podbean.com
September 22, 2025 at 11:45 PM
New Inference Time Tactics episode: Drag, Drop, and Deploy
Rob May, Calvin Cooper, Byron Galbraith, and Dave Rauchwerk share updates on the NeuroMetric Inference Time Compute Studio and why AI is moving from scaling single models to full AI system orchestration.
inferencetimetactics.podbean.com
Rob May, Calvin Cooper, Byron Galbraith, and Dave Rauchwerk share updates on the NeuroMetric Inference Time Compute Studio and why AI is moving from scaling single models to full AI system orchestration.
inferencetimetactics.podbean.com
In Ep. 5 of Inference Time Tactics, @robmay-nyc.bsky.social , Calvin Cooper, and @intrinsicmode.com unpack Salesforce’s CRMArena-Pro, going beyond “vibe testing”, and what to expect from the NeuroMetric ITC Test Engine.
Listen: inferencetimetactics.podbean.com
Listen: inferencetimetactics.podbean.com
Inference Time Tactics | NeuroMetric AI
A podcast exploring the emerging field of inference-time compute—the next frontier in AI performance. Hosted by the Neurometric team, we unpack how models reason, make decisions, and perform at runtim...
inferencetimetactics.podbean.com
September 8, 2025 at 5:21 PM
In Ep. 5 of Inference Time Tactics, @robmay-nyc.bsky.social , Calvin Cooper, and @intrinsicmode.com unpack Salesforce’s CRMArena-Pro, going beyond “vibe testing”, and what to expect from the NeuroMetric ITC Test Engine.
Listen: inferencetimetactics.podbean.com
Listen: inferencetimetactics.podbean.com
New Episode on Inference Time Tactics: Inference Time Compute Just Went Mainstream with GPT-5.
Rob May and Calvin Cooper unpack GPT-5.0’s launch and what OpenAI’s routing layer signals about the shifting AI landscape.
Listen: inferencetimetactics.podbean.com?utm_campaign... @robmay-nyc.bsky.social
Rob May and Calvin Cooper unpack GPT-5.0’s launch and what OpenAI’s routing layer signals about the shifting AI landscape.
Listen: inferencetimetactics.podbean.com?utm_campaign... @robmay-nyc.bsky.social
Inference Time Tactics | NeuroMetric AI
A podcast exploring the emerging field of inference-time compute—the next frontier in AI performance. Hosted by the Neurometric team, we unpack how models reason, make decisions, and perform at runtim...
inferencetimetactics.podbean.com
September 2, 2025 at 2:33 PM
New Episode on Inference Time Tactics: Inference Time Compute Just Went Mainstream with GPT-5.
Rob May and Calvin Cooper unpack GPT-5.0’s launch and what OpenAI’s routing layer signals about the shifting AI landscape.
Listen: inferencetimetactics.podbean.com?utm_campaign... @robmay-nyc.bsky.social
Rob May and Calvin Cooper unpack GPT-5.0’s launch and what OpenAI’s routing layer signals about the shifting AI landscape.
Listen: inferencetimetactics.podbean.com?utm_campaign... @robmay-nyc.bsky.social
95% of AI projects fail. Not from weak models, but from implementation chaos.
Next week we’re launching a preview of our AI Evaluation Platform to make deployments reliable and ROI-driven.
Learn more: open.substack.com/pub/neuromet...
Next week we’re launching a preview of our AI Evaluation Platform to make deployments reliable and ROI-driven.
Learn more: open.substack.com/pub/neuromet...
Solving the AI Implementation Crisis: Introducing NeuroMetric's Evaluation Platform
NeuroMetric is launching a preview of our evaluation platform next week. Visit https://www.neurometric.ai/ and sign up to be among the first to transform your AI implementation strategy.
open.substack.com
August 29, 2025 at 9:21 PM
95% of AI projects fail. Not from weak models, but from implementation chaos.
Next week we’re launching a preview of our AI Evaluation Platform to make deployments reliable and ROI-driven.
Learn more: open.substack.com/pub/neuromet...
Next week we’re launching a preview of our AI Evaluation Platform to make deployments reliable and ROI-driven.
Learn more: open.substack.com/pub/neuromet...
Episode 3 of Inference Time Tactics is live: When AI Overthinks:
@robmay-nyc.bsky.social, Calvin Cooper, and Byron Galbraith break down Apple’s paper, why reasoning models loop, and what it takes to make them reliable.
Listen: inferencetimetactics.podbean.com?utm_campaign...
@robmay-nyc.bsky.social, Calvin Cooper, and Byron Galbraith break down Apple’s paper, why reasoning models loop, and what it takes to make them reliable.
Listen: inferencetimetactics.podbean.com?utm_campaign...
Inference Time Tactics | NeuroMetric AI
A podcast exploring the emerging field of inference-time compute—the next frontier in AI performance. Hosted by the Neurometric team, we unpack how models reason, make decisions, and perform at runtim...
inferencetimetactics.podbean.com
August 20, 2025 at 1:46 AM
Episode 3 of Inference Time Tactics is live: When AI Overthinks:
@robmay-nyc.bsky.social, Calvin Cooper, and Byron Galbraith break down Apple’s paper, why reasoning models loop, and what it takes to make them reliable.
Listen: inferencetimetactics.podbean.com?utm_campaign...
@robmay-nyc.bsky.social, Calvin Cooper, and Byron Galbraith break down Apple’s paper, why reasoning models loop, and what it takes to make them reliable.
Listen: inferencetimetactics.podbean.com?utm_campaign...
. @robmay-nyc.bsky.social and Calvin Cooper dig into the strategic trade-offs in inference time compute — from closed vs open stacks to cost-performance decisions — and why configuration is becoming a competitive edge.
Listen here: inferencetimetactics.podbean.com
Listen here: inferencetimetactics.podbean.com
Inference Time Tactics | NeuroMetric AI
A podcast exploring the emerging field of inference-time compute—the next frontier in AI performance. Hosted by the Neurometric team, we unpack how models reason, make decisions, and perform at runtim...
inferencetimetactics.podbean.com
August 13, 2025 at 4:40 PM
. @robmay-nyc.bsky.social and Calvin Cooper dig into the strategic trade-offs in inference time compute — from closed vs open stacks to cost-performance decisions — and why configuration is becoming a competitive edge.
Listen here: inferencetimetactics.podbean.com
Listen here: inferencetimetactics.podbean.com
The next frontier in AI isn’t just training larger models. It’s getting them to reason reliably at runtime.
Inference Time Tactics is our new podcast on inference-time compute
Listen & subscribe, available on all platforms: InferenceTimeTactics.podbean.com/e/episode-1/
Inference Time Tactics is our new podcast on inference-time compute
Listen & subscribe, available on all platforms: InferenceTimeTactics.podbean.com/e/episode-1/
Why Inference Time Compute Is the Future of AI | Inference Time Tactics
Welcome to the very first episode of Inference Time Tactics — the podcast for builders, researchers, and engineers pushing the limits of AI performance.
In this kickoff conversation, hosts Rob May and...
InferenceTimeTactics.podbean.com
August 1, 2025 at 10:17 PM
The next frontier in AI isn’t just training larger models. It’s getting them to reason reliably at runtime.
Inference Time Tactics is our new podcast on inference-time compute
Listen & subscribe, available on all platforms: InferenceTimeTactics.podbean.com/e/episode-1/
Inference Time Tactics is our new podcast on inference-time compute
Listen & subscribe, available on all platforms: InferenceTimeTactics.podbean.com/e/episode-1/
I Read 200 Test Time Compute Paper Abstracts. Here Is What I Learned.
open.substack.com/pub/neuromet...
open.substack.com/pub/neuromet...
I Read 200 Test Time Compute Paper Abstracts. Here Is What I Learned.
The problems and opportunities of the third scaling law
open.substack.com
July 2, 2025 at 10:18 PM
I Read 200 Test Time Compute Paper Abstracts. Here Is What I Learned.
open.substack.com/pub/neuromet...
open.substack.com/pub/neuromet...
Benchmarking AI systems is a challenge. Figuring out the right system design is difficult. Today we release this ebook to show you how. www.neurometric.ai/ebook-challe...
April 1, 2025 at 2:12 PM
Benchmarking AI systems is a challenge. Figuring out the right system design is difficult. Today we release this ebook to show you how. www.neurometric.ai/ebook-challe...
*A Brief Intro to DataFlow Architectures* Dataflow architecture crushes the Von Neumann bottleneck, unlocking massive parallelism for AI—meet the future with Sambanova. - Read more -> open.substack.com/pub/neuromet...
February 25, 2025 at 7:27 PM
*A Brief Intro to DataFlow Architectures* Dataflow architecture crushes the Von Neumann bottleneck, unlocking massive parallelism for AI—meet the future with Sambanova. - Read more -> open.substack.com/pub/neuromet...
Last week Perplexity and Cerebras announced that they collaborated on a high performance AI search product. This is important to us at Neurometric because it validates the core idea behind the company… open.substack.com/pub/neuromet...
Perplexity and Cerebras
The first major example of hybrid AI architecture
open.substack.com
February 20, 2025 at 8:44 PM
Last week Perplexity and Cerebras announced that they collaborated on a high performance AI search product. This is important to us at Neurometric because it validates the core idea behind the company… open.substack.com/pub/neuromet...
Welcome to Neurometric
open.substack.com/pub/neuromet...
open.substack.com/pub/neuromet...
Welcome to Neurometric
We help you optimize your AI hardware for price, performance, and power.
open.substack.com
February 4, 2025 at 2:49 PM
Welcome to Neurometric
open.substack.com/pub/neuromet...
open.substack.com/pub/neuromet...
Who to follow in the Artificial Intelligence space here on BlueSky: bsky.app/profile/did:...
February 4, 2025 at 2:48 PM
Who to follow in the Artificial Intelligence space here on BlueSky: bsky.app/profile/did:...
Neurometric doesn’t have a twitter. As a new company we are focused on Bluesky as the future. To that end, we have created a list of who to follow in the AI hardware space. It’s a combination of people at NVIDIA and AMD, but also startups like Sambanova and Cerebras. List: bsky.app/profile/did:...
Our Hardware BlueSky List
Who to follow if you are into AI Hardware
open.substack.com
February 4, 2025 at 2:44 PM
Neurometric doesn’t have a twitter. As a new company we are focused on Bluesky as the future. To that end, we have created a list of who to follow in the AI hardware space. It’s a combination of people at NVIDIA and AMD, but also startups like Sambanova and Cerebras. List: bsky.app/profile/did:...
Deepseek, Shorting NVIDIA, And The Coming AI Hardware Revolution open.substack.com/pub/neuromet...
January 28, 2025 at 9:08 PM
Deepseek, Shorting NVIDIA, And The Coming AI Hardware Revolution open.substack.com/pub/neuromet...
The launch of Deepseek, an AI model with top tier performance trained at a rock bottom price, and then this very popular post (youtubetranscriptoptimizer.com/blog/05_the_...) on the short case for NVIDIA stock, have caused the tech world to suddenly re-evaluate everything. Read more on our Substack.
The Short Case for Nvidia Stock
All the reasons why Nvidia will have a very hard time living up to the currently lofty expectations of the market.
youtubetranscriptoptimizer.com
January 28, 2025 at 9:04 PM
The launch of Deepseek, an AI model with top tier performance trained at a rock bottom price, and then this very popular post (youtubetranscriptoptimizer.com/blog/05_the_...) on the short case for NVIDIA stock, have caused the tech world to suddenly re-evaluate everything. Read more on our Substack.