Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Quick Primer on MCP Using Ollama and LangChain

polarsparc.com

Post reply on HN
Print view View original

11–20 of 22 posts

Re: Quick Primer on MCP Using Ollama and LangChain

#12

Post by sunpazed »

gsibble wrote: post #2

MCP is great for when you’re integrating tools locally into IDEs and such. It’s a terrible standard for building more robust applications with multi-user support. Security and authentication are completely lacking. 99% of people wouldn’t be able to find the API keys you need to feed into most MCP servers.

While I’m a fan, we’re not using MCP for any production workloads for these very reasons.

Authentication, session management, etc, should be handled outside of the standard, and outside of the LLM flow entirely.

I recently mused on these here; https://github.com/sunpazed/agent-mcp/blob/master/mcp-what-i...

Re: Quick Primer on MCP Using Ollama and LangChain

#13

Post by lianlianlian »

bswamina wrote: post #5
gsibble wrote: post #2

MCP is great for when you’re integrating tools locally into IDEs and such. It’s a terrible standard for building more robust applications with multi-user support. Security and authentication are completely lacking. 99% of people wouldn’t be able to find the API keys you need to feed into most MCP servers.

You are correct ... it is still early days IMHO ... will have to see how this evolves

[dead]

Re: Quick Primer on MCP Using Ollama and LangChain

#14

Post by jsemrau »

bongodongobob wrote: post #11

Is anyone really still using langchain? Has it gotten better? Seemed like a token burning platform the last time I used it.

I recently finished a Langgraph class on Deeplearning.ai about a week after it came out. Already then the provided Notebook example didn't work and I needed to debug it to pass. I had great hopes on Langchain in 2024, but their product decisions toward LCEL and the complete lack of a discernible roadmap that does not constantly break things made me move away from them.

Re: Quick Primer on MCP Using Ollama and LangChain

#16

Post by minimaxir »

jsemrau wrote: post #14
bongodongobob wrote: post #11

Is anyone really still using langchain? Has it gotten better? Seemed like a token burning platform the last time I used it.

I recently finished a Langgraph class on Deeplearning.ai about a week after it came out. Already then the provided Notebook example didn't work and I needed to debug it to pass. I had great hopes on Langchain in 2024, but their product decisions toward LCEL and the complete lack of a discernible roadmap that does not constantly break things made me move away from them.

I had to look up LCEL: https://python.langchain.com/docs/concepts/lcel/

What the heck? I have no idea what problem this is solving while not also creating new problems.

Re: Quick Primer on MCP Using Ollama and LangChain

#17

Post by jsemrau »

minimaxir wrote: post #16
jsemrau wrote: post #14

Earlier quoted context omitted.

I recently finished a Langgraph class on Deeplearning.ai about a week after it came out. Already then the provided Notebook example didn't work and I needed to debug it to pass. I had great hopes on Langchain in 2024, but their product decisions toward LCEL and the complete lack of a discernible roadmap that does not constantly break things made me move away from them.

I had to look up LCEL: https://python.langchain.com/docs/concepts/lcel/ What the heck? I have no idea what problem this is solving while not also creating new problems.

The worst thing about LCEL is not that it's a different coding pattern. It also creates a major break withing the documentation that can't be fixed as you now have to factor into your search all documentation with and without LCEL.

Re: Quick Primer on MCP Using Ollama and LangChain

#19

Post by mettamage »

WD-42 wrote: post #6

If you need to define and write the functions to calculate interest… what exactly is the llm bringing to the table here? I feel like I’m missing something.

I think it's the case you don't need to but if you find it necessary. Basically you're augmenting LLMs with "normal computer power" just like a human.
Post reply on HN