Neurometric AI

Built for teams shipping AI in production. Bring your use cases and questions.

🔗 Register: luma.com/qbb8w9en

Hosted by Neurometric AI
#AI #MachineLearning #SLM #MLOps

Small Language Models (SLMs) for Production Workflows · Zoom · Luma Bigger models aren’t always better for production. In this office hour, we’ll break down how Small Language Models (SLMs) can power real-world workflows with…

Bigger models aren’t always better for production.

Hosting an office hour on Small Language Models (SLMs) for Production Workflows, covering when SLMs actually outperform large models, how to route tasks to the right model, and real examples of cutting costs without sacrificing quality.

The Deep View: Conversations | #1 - How to solve the ROI problem for AI inference - Rob May How do you make AI inference affordable enough to deliver real ROI in the enterprise?In this episode of The Deep View Conversations, we talk with Rob May, founder and CEO of Neurometric AI, to brea...

Read more: tdv.transistor.fm/episodes/1-h...

YouTube video by the Deep View How to solve the ROI problem for AI inference

Enterprise AI has an ROI problem. Inference costs are skyrocketing and the brute force approach is failing.
Neurometric CEO Rob May joined The Deep View to break down how we use thinking algorithms and specialized models to drop costs without sacrificing performance. www.youtube.com/watch?v=FS7v...

We just launched the Neurometric Audio Leaderboard: leaderboard.neurometric.ai

This week’s update includes Modulate Velma 2, released yesterday.

The leaderboard tracks real-world performance of voice and audio models as the space moves toward more specialized and cost-efficient architectures.

New episode of Inference Time Tactics.

Calvin Cooper and Byron Galbraith talk with Rapt.AI about why inference workloads are hard to predict and what it takes to run them efficiently in production.

Listen now:
inferencetimetactics.podbean.com

We cover why only 20% of companies are positioned to scale AI, why cost overruns are so common, and why data quality is now the top blocker in production systems.
Listen: inferencetimetactics.podbean.com
#EnterpriseAI #GenAI #AIOps

Episode 11 of Inference Time Tactics features Shawn Rogers, CEO of BARC, unpacking insights from new research based on 421 organizations deploying AI in production worldwide.

Explore the data: leaderboard.neurometric.ai

Behind the scenes with Calvin Cooper and Byron Galbraith on the engineering, the research, and what enterprises need to know.

Listen: inferencetimetactics.podbean.com

Neurometric launched the first leaderboard that combines models WITH thinking algorithms—not just single model performance. Our surprising finding: performance varies dramatically per task, and no single system dominates.

Leaderboard: leaderboard.neurometric.ai

Founders Everywhere: Rob May Rob May is the co-founder & CEO of Neurometric, a startup providing automated inference orchestration for multi-model AI systems.

@robmay-nyc.bsky.social featured today in Founders Everywhere by @EverywhereVC

We are building automated inference orchestration for multi-model AI systems so teams can measure, not guess, what drives real performance.

Read the spotlight: ideas.everywhere.vc/p/neurometri...

#AI #Startups

Leaderboard: leaderboard.neurometric.ai

The Deep View Join 600,000+ readers from Google, Meta, Microsoft and more who read The Deep View daily to get smarter about AI.

Our Thinking Algorithm Leaderboard is now live - the first public benchmark ranking model + reasoning algorithm combinations instead of models alone.

Covered today in The Deep View (600k+ subscribers): www.thedeepview.com/articles/ope...

#AI #LLM #AIOptimization #ThinkingAlgorithms #ITC #TTC

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

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.

Generalization is not determined by model design alone. It emerges from how systems behave at runtime.

The frontier is shifting from training optimization to inference orchestration.

Read here: open.substack.com/pub/neurometric/p/apples-to-infinity-and-beyond-shows

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.

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.

Beyond Benchmark Hype: Why Inference-Time Strategy is the Real AI Bottleneck How research from Princeton's HAL project confirms what we're seeing in production: the path to reliable AI isn't bigger models. It's smarter orchestration.

At Neurometric, we’ve seen the same in production. The next phase of AI performance isn’t about bigger models, it’s about smarter inference.

Full post by Calvin Cooper here: neurometric.substack.com/p/beyond-ben...

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

Tune in: 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.

Read more: neurometric.substack.com/p/introducing-itc-studio

Try it here: dirtbike.neurometric.xyz

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

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

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

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

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.

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...

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...