@kmeanskaran: Guys, I'm building a PoC called "Agent Harness Ops." From dev phase to prod on @Railway What's special? - Created an ag…

X AI KOLs Timeline News
agent-harness poc langchain deployment railway ci-cd

Summary

Building a PoC called 'Agent Harness Ops' using DeepAgents by LangChain, with observability via Langfuse, task queue with Celery/Redis, PostgreSQL, React frontend, deployed on Railway. Plans to share and write about it.

Guys, I'm building a PoC called "Agent Harness Ops." From dev phase to prod on @Railway What's special? - Created an agent harness using DeepAgents by LangChain - Plugged in Langfuse for observability - Using Celery workers and Redis for the task queue - Saving users' historical generations in PostgreSQL - Running subagents in parallel, with human-in-the-loop for approval - React JS as the frontend, with an NGINX reverse proxy - Deploying FastAPI (backend), Celery workers, Redis, PostgreSQL, and React JS on Railway Syncing it all together with CI/CD and scaling. I personally love Railway and wanted to showcase something, but this time I'm building a complete Ops cycle instead of just an agent. I'll be sharing this soon and writing about it, so you can also deploy your agents on Railway at lower cost and with a simpler setup than AWS/GCP/Azure. There are tradeoffs, which I'll discuss later. Also I will post about the agent harness dev phase first and then about production.
Original Article
View Cached Full Text

Cached at: 07/03/26, 04:41 PM

Guys, I’m building a PoC called “Agent Harness Ops.” From dev phase to prod on @Railway

What’s special?

  • Created an agent harness using DeepAgents by LangChain
  • Plugged in Langfuse for observability
  • Using Celery workers and Redis for the task queue
  • Saving users’ historical generations in PostgreSQL
  • Running subagents in parallel, with human-in-the-loop for approval
  • React JS as the frontend, with an NGINX reverse proxy
  • Deploying FastAPI (backend), Celery workers, Redis, PostgreSQL, and React JS on Railway

Syncing it all together with CI/CD and scaling. I personally love Railway and wanted to showcase something, but this time I’m building a complete Ops cycle instead of just an agent.

I’ll be sharing this soon and writing about it, so you can also deploy your agents on Railway at lower cost and with a simpler setup than AWS/GCP/Azure.

There are tradeoffs, which I’ll discuss later. Also I will post about the agent harness dev phase first and then about production.

Similar Articles

@Potatoloogs: https://x.com/Potatoloogs/status/2057391224592667051

X AI KOLs Timeline

This article deeply analyzes the concept of Agent Harness, which is the engineering infrastructure wrapped around an LLM, including 12 components such as orchestration loops, tool calling, memory systems, context management, etc. The article cites practices from companies like Anthropic, OpenAI, and LangChain, arguing for the critical role of the harness in production-grade AI agents.