Kern
DocsAgentsWorkflowsTools
Get Started

Engineering Blog

Notes from the reliability layer

Playbooks, post-mortems, and engineering insights on structured data synthesis and workflow execution with small language models.

Why small models fail at structured output
Reliability
May 12, 2026•8 min read

Why small models fail at structured output

An in-depth analysis of why sub-7B models struggle with strict JSON schemas, and how to fix it with templates, JSON repair, and validation loops.

Read Article
Making local LLMs production reliable
Deployment
May 5, 2026•9 min read

Making local LLMs production reliable

A comprehensive playbook for deploying llama.cpp, Ollama, and Kern in production and regulated environments.

Read Article
Benchmarking AI workflows
Benchmarks
April 24, 2026•10 min read

Benchmarking AI workflows

Why single-prompt metrics fail to represent agent performance, and how to measure multi-step system success.

Read Article
AI reliability infrastructure
Thesis
April 12, 2026•6 min read

AI reliability infrastructure

Why reliability is the missing layer in the modern AI stack, and why application code shouldn't handle non-deterministic errors.

Read Article
Observability for AI agents
Observability
March 30, 2026•8 min read

Observability for AI agents

Design patterns for tracing non-deterministic agent workflows, and the metrics SREs should monitor.

Read Article
Kern

Small Models. Big Impact.

An agent framework built for 1-7B parameter LLMs.

Documentation

  • Getting Started
  • Agents
  • Workflows
  • Tools
  • Models

Resources

  • Structured Output
  • Visualization
  • Knowledge Base
  • Storage

Community

  • GitHub
  • Discord

© 2026 NDX Rocks. Apache 2.0 License.