DEV Community

Machine Learning

A branch of artificial intelligence (AI) and computer science which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy.

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
How I Cut My Client's Image Analysis Costs by 90% — A Multimodal API Showdown for 2026

How I Cut My Client's Image Analysis Costs by 90% — A Multimodal API Showdown for 2026

Comments
9 min read
What poisoning a RAG store taught us about agent memory

What poisoning a RAG store taught us about agent memory

1
Comments 2
7 min read
Stop Guessing: Real p99 Latency Data Comparing DeepSeek, Qwen, Kimi, and GLM

Stop Guessing: Real p99 Latency Data Comparing DeepSeek, Qwen, Kimi, and GLM

Comments
8 min read
Verifiable rewards improve LLM math accuracy

Verifiable rewards improve LLM math accuracy

Comments
3 min read
The Developer's Guide to Breaking Free from Proprietary Multimodal AI

The Developer's Guide to Breaking Free from Proprietary Multimodal AI

Comments
8 min read
Porting a 1,200-line persistent CUDA megakernel to Qwen3-TTS: ~25 ms to first audio chunk

Porting a 1,200-line persistent CUDA megakernel to Qwen3-TTS: ~25 ms to first audio chunk

2
Comments
3 min read
Stop Guessing: Real Data Comparing US and Chinese AI Models in 2026

Stop Guessing: Real Data Comparing US and Chinese AI Models in 2026

Comments
5 min read
Your Provenance Vector Dies at the Storage Boundary

Type-level gates and memory compaction

Your Provenance Vector Dies at the Storage Boundary

14
Comments 18
8 min read
DeepSeek vs Qwen vs Kimi vs GLM: Which Chinese AI Model Actually Wins in 2026?

DeepSeek vs Qwen vs Kimi vs GLM: Which Chinese AI Model Actually Wins in 2026?

Comments
9 min read
I Wish I Knew This Speed Hack Sooner — Here's the Full Breakdown

I Wish I Knew This Speed Hack Sooner — Here's the Full Breakdown

1
Comments
6 min read
ArchGuard: Detect Architecture Drift Before It Becomes Technical Debt

ArchGuard: Detect Architecture Drift Before It Becomes Technical Debt

2
Comments
6 min read
The SLM-First Agent: Why 2026's Best Agentic Systems Run on Small Models

The SLM-First Agent: Why 2026's Best Agentic Systems Run on Small Models

Comments
6 min read
Our LiDAR detector spent 40% of its time in voxelization, not convs

Our LiDAR detector spent 40% of its time in voxelization, not convs

1
Comments
4 min read
Five things I noticed this week: GLM-5.2, Noam Shazeer at OpenAI, and 10k malware repos

Five things I noticed this week: GLM-5.2, Noam Shazeer at OpenAI, and 10k malware repos

1
Comments
4 min read
I Tested DeepSeek V4 Flash and GPT-4o Side by Side — Here's the Real-World Performance Data

I Tested DeepSeek V4 Flash and GPT-4o Side by Side — Here's the Real-World Performance Data

7
Comments 2
6 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.