Eat Sleep Append Repeat…
Re-architecting LangWatch on event sourcing and ClickHouse hit 6,000 events per second, real-time dashboards, and retroactive data improvements via replay.
LangWatch Blog6 min read
LangWatch
30 stories
Four Refactors and a Funeral: Migrating a Live System to Event Sourcing
How LangWatch migrated a live platform to event sourcing on ClickHouse: four refactors, a fold-map-react pipeline, and a zero-downtime dual-write cutover.
LangWatch Blog20 min read
Internal Product vs Internalised Trauma: Supporting Event Sourced Systems
The tooling that keeps an event-sourced system supportable: Grafana dashboards, an ops console for blocked groups, projection replay, and time-travel debugging.
LangWatch Blog7 min readEvery way your AI agent can be broken (and how attackers actually do it)
A practical AI agent red teaming guide: attacker goals, the Crescendo multi-turn strategy, OWASP LLM Top 10 mapping, and how to test agents before shipping.
LangWatch Blog11 min readWhy AI Red teaming is broken (and how we fixed it)
Single-turn AI red teaming misses real attacks. See how multi-turn Crescendo simulations in Scenario break agents that pass every benchmark.
LangWatch Blog9 min read
How we test Agent Skills with Scenario simulations
Most AI agent skills ship untested. See how LangWatch Scenario simulates full multi-turn conversations with a user simulator and judge to catch real failures.
LangWatch Blog7 min readGetting to value with LangWatch, faster than ever
Set up evals, agent simulations, and migrate from Langfuse or LangSmith to LangWatch in one session using Skills and MCP, not a full sprint of work.
LangWatch Blog5 min readA Note on the LiteLLM Vulnerability
A malicious credential-stealing file appeared in LiteLLM 1.82.8. LangWatch customers were not impacted because all dependencies are pinned with lockfiles.
LangWatch Blog2 min readProduct Managers and leaders are running agent simulations now, and it changing how AI ships
Non-technical leaders now write agent simulations in plain language with LangWatch and Claude, testing AI agents before a single customer sees them.
LangWatch Blog6 min readMaking your AI Agent reliable: Adding Evaluations to your multi-modal agent with LangWatch Skills
Add evaluations to a multimodal AI agent with LangWatch Skills, scaffolding experiments, tool-usage checks, and Scenario simulations in about 30 minutes.
LangWatch Blog9 min read
LangWatch Skills: Your coding agent already knows how to test your agent
LangWatch Skills let your coding agent instrument tracing, observe production performance, and run tests automatically, no manual setup needed.
LangWatch Blog6 min readIntroducing LangWatch MCP: Test and evaluate AI Agents without leaving your workflow
The LangWatch MCP server brings agent evaluation into Claude, Cursor, and Copilot, so you version prompts, generate scenarios, and benchmark models in flow.
LangWatch Blog6 min read
Showing 30 of 30 stories
You're all caught up.
AiByts Weekly Brief
Get the top AI news and tools straight to your inbox every week.
Weekly · AI news
Join 15,000+ AI enthusiasts