Inspiration
I have been working in compliance industry for a few years now. Managing compliance is very hard and is tougher when it has to be done from start. Starting my last company I had no right team to do it and all the tools were very expensive, non-context aware. I wanted to build a product that would help startups get compliant faster. With the technology we have today, it just became possible to do so.
What it does
ComplianceThunder is a context aware platform that will analyze everything you have about your organization whether from your website or existing policies. It will analyze the gaps and then provide you the policies and procedures that you require to get compliant. It reduces months of work for multiple team members to minutes.
How I built it
It is built with help of code generation, using multiple platforms like ApiFy, Anthropic, AWS, Operant AI. I started with building an agent that would crawl websites and find relevant information about the organization, suggesting the right compliance for them, then when users uploads document of their existing policies, it finds the gaps and generates the policies.
Challenges we ran into
Had lots of issues with integrating AWS Bedrock, Operant and other tools. Specially working alone was a bit difficult to do.
Accomplishments that I am proud of
It works end to end. Reduces time to get gap analysis to very short which in today's world consultants charge heavily. This also ensures that the document uploaded by the user shouldn't contain any sensitive information that may leak to LLMs.
What we learned
AI agents can solve lot of monotonous burden task even in tech field.
What's next for ComplianceThunder
Extend the product to end to end competing with old school compliance automation tools.
Built With
- amazon-web-services
- anthropic
- apify
- fastapi
- operant
- python
- stremlit
Log in or sign up for Devpost to join the conversation.