DEV Community

Multigrid profile picture

Multigrid

Building Multigrid — one API and one balance across every major model provider, with routing, failover and cost telemetry. I write about inference: what it costs, why it breaks, how to measure it.

Joined Joined on  Personal website https://multigrid.ai

Work

Building Multigrid (multigrid.ai)

Evaluating Classical NLP vs LLM Approaches

Evaluating Classical NLP vs LLM Approaches

Comments
5 min read

Want to connect with Multigrid?

Create an account to connect with Multigrid. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
Task-Specific Metrics: BLEU, ROUGE, and Their Limits

Task-Specific Metrics: BLEU, ROUGE, and Their Limits

5
Comments
5 min read
A Next.js Route Handler That Calls a Model

A Next.js Route Handler That Calls a Model

Comments
5 min read
Cadence, Volume and the Slop Problem

Cadence, Volume and the Slop Problem

Comments
7 min read
Neuron-Level Explanations and Their Limits

Neuron-Level Explanations and Their Limits

Comments
6 min read
Nested and Recursive Schemas: The Depth Limit Nobody Documents

Nested and Recursive Schemas: The Depth Limit Nobody Documents

Comments
5 min read
Neural Radiance Fields and Gaussian Splatting

Neural Radiance Fields and Gaussian Splatting

Comments
7 min read
Negative Prompts and How They Work Mechanically

Negative Prompts and How They Work Mechanically

Comments
5 min read
Negative Instructions: Why "Don't" Often Backfires

Negative Instructions: Why "Don't" Often Backfires

Comments
4 min read
Named Entity Recognition Then and Now

Named Entity Recognition Then and Now

Comments
5 min read
Music and Sound Generation Models

Music and Sound Generation Models

Comments
6 min read
Multimodal RAG: Retrieving Over Images and Text

Multimodal RAG: Retrieving Over Images and Text

Comments
4 min read
Multilingual Search Without Separate Indexes

Multilingual Search Without Separate Indexes

Comments
5 min read
Multilingual Embeddings: Cross-Language Retrieval

Multilingual Embeddings: Cross-Language Retrieval

Comments
5 min read
Multi-Tenant RAG Without Leaking Between Customers

Multi-Tenant RAG Without Leaking Between Customers

Comments
5 min read
Multi-Region AI Serving

Multi-Region AI Serving

Comments
7 min read
Multi-LoRA Serving: Many Adapters, One Base Model

Multi-LoRA Serving: Many Adapters, One Base Model

Comments
5 min read
Multi-GPU Inference: Tensor and Pipeline Parallelism

Multi-GPU Inference: Tensor and Pipeline Parallelism

Comments
5 min read
Multi-Armed Bandits for Product Decisions

Multi-Armed Bandits for Product Decisions

Comments
7 min read
Multi-Agent Systems: When Two Agents Beat One

Multi-Agent Systems: When Two Agents Beat One

Comments
5 min read
Multi-Agent Context: What Each Agent Should See

Multi-Agent Context: What Each Agent Should See

Comments
5 min read
MTEB and How Embedding Models Are Ranked

MTEB and How Embedding Models Are Ranked

Comments
6 min read
Currency, Rounding and Why Money Should Be Integers

Currency, Rounding and Why Money Should Be Integers

Comments
5 min read
Model Weights, Checkpoints and What “Open” Really Means

Model Weights, Checkpoints and What “Open” Really Means

Comments
5 min read
Model Weights in CI/CD

Model Weights in CI/CD

Comments
6 min read
Supply Chain Risk in Open Model Weights

Supply Chain Risk in Open Model Weights

Comments
4 min read
Pruning, Sparsity, and What the Hardware Can Exploit

Pruning, Sparsity, and What the Hardware Can Exploit

Comments
7 min read
Model Parameters: What 7B, 70B and 400B Actually Buy You

Model Parameters: What 7B, 70B and 400B Actually Buy You

Comments
4 min read
“Model Not Found” and the Naming Traps Behind It

“Model Not Found” and the Naming Traps Behind It

Comments
5 min read
How Long a Model Stays Current: Measuring Deprecation Properly

How Long a Model Stays Current: Measuring Deprecation Properly

Comments
5 min read
Model Interpretability: SHAP, LIME and Their Limits

Model Interpretability: SHAP, LIME and Their Limits

Comments
4 min read
How Big a Model Is on Disk: Parameters Times Bytes Per Weight

How Big a Model Is on Disk: Parameters Times Bytes Per Weight

Comments
5 min read
GGUF, Safetensors and Model File Formats

GGUF, Safetensors and Model File Formats

Comments
5 min read
Feature Flags for Models and Prompts

Feature Flags for Models and Prompts

Comments
6 min read
Model Extraction and Distillation Attacks

Model Extraction and Distillation Attacks

Comments
4 min read
Model Editing: Changing One Fact

Model Editing: Changing One Fact

Comments
5 min read
Model Drift and When to Retrain

Model Drift and When to Retrain

Comments
6 min read
Distillation: Teaching a Small Model From a Big One

Distillation: Teaching a Small Model From a Big One

Comments
5 min read
Canary and Blue-Green Deploys for Model Changes

Canary and Blue-Green Deploys for Model Changes

Comments
6 min read
Model Degradation Over Time: Real or Perceived?

Model Degradation Over Time: Real or Perceived?

5
Comments
4 min read
MCP: The Model Context Protocol, Explained

MCP: The Model Context Protocol, Explained

Comments
6 min read
The Commoditisation of Model Capability

The Commoditisation of Model Capability

Comments
6 min read
Model Collapse: What the Research Actually Showed

Model Collapse: What the Research Actually Showed

Comments
5 min read
Model Cascading: Cheap Model First, Expensive on Failure

Model Cascading: Cheap Model First, Expensive on Failure

Comments
5 min read
Model Cards and System Cards: Reading Them Critically

Model Cards and System Cards: Reading Them Critically

Comments
4 min read
Canary Releases for Model Migrations

Canary Releases for Model Migrations

Comments
6 min read
Calibration: Making 0.8 Mean 80%

Calibration: Making 0.8 Mean 80%

Comments
8 min read
MMLU: What It Tests and What a Score Means

MMLU: What It Tests and What a Score Means

Comments
7 min read
Machine Learning vs Deep Learning vs AI: Three Nested Sets

Machine Learning vs Deep Learning vs AI: Three Nested Sets

Comments
4 min read
A/B Testing Machine Learning Systems

A/B Testing Machine Learning Systems

Comments
6 min read
Mixture of Experts: Why a 400B Model Can Cost Like a 40B One

Mixture of Experts: Why a 400B Model Can Cost Like a 40B One

Comments
2 min read
The Metrics That Matter: A Minimal LLM Dashboard

The Metrics That Matter: A Minimal LLM Dashboard

Comments
5 min read
What a 1M-Token Context Window Is Actually Good For

What a 1M-Token Context Window Is Actually Good For

Comments
4 min read
Meta-Prompting: Using an LLM to Write Your Prompts

Meta-Prompting: Using an LLM to Write Your Prompts

Comments
5 min read
Mentoring Juniors in an AI-Assisted Team

Mentoring Juniors in an AI-Assisted Team

Comments
5 min read
Memory Bandwidth Is the Real Bottleneck

Memory Bandwidth Is the Real Bottleneck

Comments
5 min read
Memorisation and Generalisation, Measured

Memorisation and Generalisation, Measured

Comments
5 min read
Mechanistic Interpretability: Reading a Model's Mind

Mechanistic Interpretability: Reading a Model's Mind

Comments
5 min read
Measuring Hallucination Rate in Your Own App

Measuring Hallucination Rate in Your Own App

Comments
5 min read
Monte Carlo Tree Search for LLM Reasoning

Monte Carlo Tree Search for LLM Reasoning

Comments
5 min read
loading...