» Tag
machinelearning
91 postsAI in SRE: A Practical Autonomy Model for Self-Healing Infrastructure
An exploration of AI's role in SRE and the importance of self-healing infrastructure.
SkewAdam: A New Optimizer Reducing MoE State Memory by 97%
SkewAdam reduces MoE state memory by 97%, allowing a 6.7B model to fit on a 40GB GPU.
I Trained a 30M-Parameter LLM from Scratch: The Scaling "Floor" Was a Mirage
A 30M parameter language model was trained from scratch, aiming to deepen understanding of transformers.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comIs AI Progress Real? Four Independent Metrics Show It
Four independent metrics reveal significant AI performance increases at the end of 2025.
AI Society Developed to Debate Outages is Three Times More Reliable
A multi-agent system using Qwen shows three times the reliability of a single agent in incident response.
Beyond Scaling Laws: The Importance of Thinking Longer in AI
The significance of test-time compute in AI model development is increasing.
When AI Can't Read, It Invents — But It Still Sees the Shape
Latest AI models struggle with low-resolution document reading, risking accounting errors with fabricated details.
Building a Multi-Horizon Crypto Prediction System with Transformers
Explore the engineering story of crypto prediction with Transformers. Lessons on transparency and reliability with real metrics.
TokenRouter: A Serving Engine for Token-Level LLM Routing
TokenRouter serves as a token-level LLM routing engine, enabling collaboration among models.
Read in Parallel, Reason in Depth for Long-Context LLM Agents
PARSER improves long-context processing by enabling parallel reading and deep reasoning, enhancing accuracy and reducing latency.