kaushik92
Show HN: Compliant LLM toolkit for ensuring compliance & security of AI systems
https://github.com/fiddlecube/compliant-llm
With the right technique, I was able to break the so-called secure models like Claude and OpenAI.So, I built an open-source tool to automate this and find security holes in any hosted model.I got claude-sonnet-4 to demonstrate the following harmful behavior:- steal data from downstream tool calls using sql injection, code injection and template injection attacks- install spyware or malware...
Show HN: Compliant-LLM: Audit AI Agents for Compliance with NIST AI RMF
https://github.com/fiddlecube/compliant-llm
We're excited to launch compliant-llm: an open-source toolkit that helps infosec and compliance teams audit AI agents against regulatory frameworks like NIST AI RMF, ISO 42001, and OWASP Top 10.Infosec and compliance teams are now responsible for tracking security and compliance risks of a growing number of AI agents across external and internal apps and third-party vendors.compliant-llm gives...
Re:Show HN: FiddleCube – Generate Q&A to test your LLM
on Friday June 28, 2024 @03:51PM (#40824558)
I have seen this, but I find it fairly hard to use.
Our goal is to focus on datasets and make it very easy to create and manage data.
In our next release, we will be launching a way to do this using a UI.
Re:Show HN: FiddleCube – Generate Q&A to test your LLM
on Friday June 28, 2024 @03:45PM (#40824500)
We identified and solved for 2 key problems with generating data using GPT:
1. Duplicate/similar data points - we solve this by adding deduplication to our pipeline.
2. Incorrect question-answers - we check for correctness and context relevance. Filter out incorrect rows of data.
Apart from this, we generate a diverse set of questions including complex reasoning and chain of thought.
We also generate domain specific unsafe questions - questions that violate TnC of the particular LLM to test the model guardrails.
Re:Show HN: FiddleCube – Generate Q&A to test your LLM
on Wednesday June 26, 2024 @11:50AM (#40801381)
This is great feedback! Will work on adding this soon.
Re:Show HN: FiddleCube – Generate Q&A to test your LLM
on Wednesday June 26, 2024 @11:48AM (#40801362)
We do care a lot about data privacy. While the data is sent to an API, we do not store anything on our servers or use our user's data in any way.
We are working on getting SOC2 certified. In the meantime, we sign a legally binding agreement with our users who have data privacy needs/concerns.
Re:Show HN: FiddleCube – Generate Q&A to test your LLM
on Tuesday June 25, 2024 @02:53PM (#40792116)
Ragas is an eval tool which needs ground truths and queries for evaluation. FiddleCube generates the queries and the ground truth needed for eval in Ragas, LangSmith or an eval tool of choice.
We incorporate user prompts to generate the outputs and provide diagnostics and feedback for improvement, rather than eval metrics. So you can plug your low scored queries provided by Ragas, your prompt and context. FiddleCube can provide the root cause and the ideal response.
This is an alternative to manual auditing and testing, where an auditor works on curating the ideal dataset.
Show HN: FiddleCube – Generate Q&A to test your LLM
https://github.com/fiddlecube/fiddlecube-sdk
Convert your vector embeddings into a set of questions and their ideal responses. Use this dataset to test your LLM and catch failures caused by prompt or RAG updates.Get started in 3 lines of code:```pip3 install fiddlecube``````from fiddlecube import FiddleCubefc = FiddleCube(api_key="<api-key>") dataset = fc.generate( [ "The cat did not want to be petted.", "The cat...
Re:Show HN: A lightweight AI gateway to 100+ models, in TS
on Monday January 08, 2024 @01:33PM (#38916081)
Congrats on the launch! Looks amazing
Re:Show HN: Fully automated codebase migrations and upgrades using AI
on Monday August 28, 2023 @12:30PM (#37296675)
Migration projects are painful and no developer enjoys them.
Good to see a great solution to automate this!
Re:Show HN: Litellm – simple library to standardize OpenAI, Cohere, Az...
on Thursday July 27, 2023 @12:38PM (#36896206)
This is amazing. Really needed something like this to standardize all my different AI APIs!
On a side note - I love how quickly your team is shipping! Do keep it going!
Re:Launch HN: PeerDB (YC S23) – Fast, Native ETL/ELT for Postgres
on Thursday July 27, 2023 @12:28PM (#36896053)
Moving data in and out of Postgres in a fast and reliable way is exactly what my startup needs. I am looking forward to trying PeerDB!