Scarlett Attensil

Factory.ai defines a software factory as a nine-stage loop: signals, triage, plan, build, test,...

This tutorial explains how to catch a dangerous agent skill before an agent ever runs it: review it automatically, block it in CI if it fails, and only let your agent load skills that passed.

Hand-tuning prompts for a subjective quality like tone does not scale. Use LaunchDarkly agent optimization to explore candidate prompts for a reply-drafting agent against GPTZero, an AI detector wired in as an inverted judge, then run a fuller eval on the candidates it surfaces. The run streams to the optimization Results tab and commits the winner as a variation.

Agent framework debates are mostly vibes. One engineer swears LangGraph is faster, another prefers...

Learn how to evaluate, experiment with, and safely roll out AI changes using metrics, guardrails, AgentControl configs, online evaluations, and LaunchDarkly release controls.

Understand the MLOps lifecycle from data preparation to deployment, monitoring, governance, and retraining.

Learn why uncontrolled AI pipeline changes can cause more failures than bad models in production RAG systems.

Every woman who has worked in tech long enough has gotten the same three pieces of advice. Confront...

Layoff years compress organizations. Scopes overlap. Lanes that were clear during a growth year start...

My daily workflow looks nothing like it did a year ago. A lot has landed in Claude Code recently....

A silent provider update once invalidated months of LLM-as-judge scores in a pipeline I owned. Here is what changed after, and why parenting taught me the same lesson twice.

Regression-test your support agent's RAG generation offline using LaunchDarkly Datasets, the Playground, and LLM-as-a-judge. What offline evals can and can't tell you.

Build a 3-agent research analysis swarm that runs across three orchestrators. Learn how LaunchDarkly AI Configs enable framework-agnostic agent definitions, per-agent cost tracking, and silent failure detection.

Install LaunchDarkly Agent Skills in your AI coding assistant and create AI Configs from natural language. Build a three-agent pipeline and wire it to LangGraph.

Build custom LLM-as-judge evaluators for AI code generation. Score security, API contracts, and scope creep. Compare models with data from your codebase.

Build a multi-agent system with dynamic configuration in 20 minutes using LangGraph multi-agent workflows, RAG search, and LaunchDarkly AI Configs.

A practical guide for LLM observability, AI evaluation, and real-time quality monitoring in production with LaunchDarkly

Understand when to use prompt-based mode vs agent mode in LaunchDarkly AI Configs. Learn the real differences, provider patterns, and decision frameworks for LangGraph, OpenAI, and multi-agent systems.

How Session Replay + Online Evals Revealed How My Holiday Pet App Actually Works Original...

Measure & prove the ROI of AI model changes with LaunchDarkly experiments. Discover unexpected performance wins and cost paradoxes via statistical analysis.

Transform your basic multi-agent system with business tiers, geographic targeting, and MCP tool integration using LaunchDarkly AI Configs.

Learn to build LangGraph multi-agent workflows with RAG search and dynamic AI configuration using LaunchDarkly AI Configs - no deployments needed for changes.