Summary
Itacon aimed to develop an AI solution for generating comprehensive risk analyses from simple URLs. We created a system that profiles companies and delivers detailed risk assessments, including legislative forms, business areas, and asset lists. By combining a pre-trained LLM model with a specialized prompt-based system, we streamlined the process with minimal inputs. The solution, enhanced by a feedback loop pipeline, has revolutionized their risk analysis capabilities, providing thorough and efficient assessments.
Challenge
Motivation to change
The goal of the project is the development of AI solution responsible for collecting information about company, and based on this generate a comprehensive risk analysis.
The input to the solution is a simple URL with following outputs:
- Risk Analysis
- Address and Legislative Form
- Business Area
- Asset List
Solution
Change delivery
The core challenge is a creation of a solution capable of understanding company profile and build custom-made risk analysis.
We also build feed-back loop pipeline to enable users intervened in the process of generating asset to improve the solution in the future.
Combination of a pre-trained LLM model with specialized prompt-based system enabled us to build entire solution with as little inputs as possible.
Tools & means
- Python
- FastAPI
- OpenAI
- LangChain
Outcomes
Change outcomes
Working with us has been a game-changer for them. Our ability to generate a comprehensive risk analysis from a simple URL is unparalleled. The thorough company profiling, asset generation, and meticulous risk assessment we provide have significantly enhanced their process.