In this episode of our podcast “High Bit,” Automat founder and CTO Gautam Bose discusses the challenges of using large language models for robotic process automation. In particular, Gautam stresses the essential need for thorough testing in automation. He points out that while initial tests might show promising results, they do not always predict long-term success and reliability, underscoring the importance of comprehensive testing strategies.

Highlights include:
– Challenges of building reliable RPA with LLMs.
– Automat’s rigorous LLM testing for sensitive data.
– Automating complex workflows via SOP collections.
– Future of RPA: human-AI collaboration, self-improving systems.
– Examples of complex automation and iterative improvements.


Automat uses large language models to rapidly automate complex business workflows at a fraction of traditional costs.

Learn more about their solution:  https://www.runautomat.com/