Runnable companion code for articles on mpiv.ai/blog covering AI agent infrastructure: LiteLLM proxy setups, agent observability with Langfuse / Braintrust / Helicone, evaluation pipelines, multi-provider gateways.
Each subdirectory corresponds to one published article. The subdirectory is created when the article first references it; code is added as needed to make the article's instructions reproducible by a reader following along.
| Article | Subdirectory | Status |
|---|---|---|
| How to Set Up LiteLLM Proxy for Production AI Agents (2026) | litellm-proxy-setup/ |
Code samples coming — open an issue if you'd like a specific snippet packaged |
| LiteLLM vs OpenRouter: Which Wins for Production AI Agents (2026) | litellm-vs-openrouter/ |
Planned |
| Production Agent Observability (2026) | production-agent-observability/ |
Planned |
| Langfuse vs Braintrust (2026) | langfuse-vs-braintrust/ |
Planned |
| Agent Eval Pipelines (2026) | agent-eval-pipelines/ |
Planned |
- One subdirectory per article. Subdirectory name is the article slug (truncated where useful).
- Each subdirectory has its own
README.mdlinking back to the article and explaining how to run the code. - Code is intended to be runnable as-is on a fresh machine with the listed prerequisites — no proprietary infra, no commercial-tier cloud accounts beyond what the article itself calls out.
- License: MIT unless a subdirectory specifies otherwise.
MIT — see LICENSE.