Agentic AI: Transforming Vulnerability Management with Harry Wetherald

Jul 31, 2025

Harry Wetherald, Co-Founder and CEO of Maze, shares his expertise in AI and machine learning, particularly in the realm of vulnerability management. He delves into the concept of agentic AI, which allows AI to independently analyze vulnerabilities, massively enhancing efficiency. The conversation highlights the critical need for context engineering to tailor AI solutions for diverse organizations. Harry also discusses the hurdles of achieving reliable AI systems and emphasizes the importance of clear pricing strategies to improve customer experience and budget predictability.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

ANECDOTE

Founders' Firsthand Frustration

Harry and co-founders lived the scaling pain of huge vulnerability backlogs in large engineering teams.
That experience drove them to replicate how an expert human analyst reasons when building Maze.

INSIGHT

Agents Replicate Human Investigations

Agents let systems investigate like humans by using tools and calling relevant APIs step-by-step.
This avoids preloading all data and allows targeted context gathering for each vulnerability.

ADVICE

Reduce Hallucination By Tight Reasoning

Tighten reasoning loops at every decision point to prevent compounding hallucinations in agents.
Combine continuous human interaction with engineering optimizations to reach enterprise-grade reliability and control costs.

Get the Snipd Podcast app to discover more snips from this episode

Get the app

In this episode, we talk to Harry Wetherald (Co-Founder and CEO @ Maze). We explore the evolving landscape of vulnerability management. Harry shares insights from his journey in AI and machine learning, discussing the challenges of triaging vulnerabilities across diverse organizations. The conversation delves into the concept of agentic AI, the importance of context engineering, and the hurdles of achieving enterprise-grade reliability in AI systems. Harry also reflects on the inflection points that led to the founding of Maze and the role of LLMs in transforming security practices.

Key Takeaways

- Introduction to Maze and Harry's Journey: Harry shares his background in AI and machine learning, emphasizing the persistent challenges in vulnerability management and the founding of Maze to address these issues.

- Agentic AI and Context Engineering: The discussion highlights the shift from static rules to agentic AI, where AI agents autonomously investigate vulnerabilities, and the critical role of context engineering in tailoring solutions to specific organizational needs.

- Challenges in AI Reliability: Harry talks about the engineering hurdles in making AI systems reliable and consistent, focusing on the importance of tight reasoning loops and human-AI symbiosis.

- Pricing Strategies: In AI-native security solutions, clear and fixed pricing is preferred, as it simplifies budgeting and aligns with traditional models, while vendors should manage costs to ensure predictability for customers.

- Future of Security with AI: The conversation concludes with insights into the future of security, where AI agents work in the background to provide innovative solutions, and the importance of human feedback in refining AI systems.

Tune in for a deep dive into the future of security and the innovative approaches shaping the industry!

Contacting Harry

* Harry's LinkedIn: https://www.linkedin.com/in/harrywetherald/

* Maze: https://mazehq.com/

Contacting Anshuman

* LinkedIn: ⁠⁠⁠⁠https://www.linkedin.com/in/anshumanbhartiya/

* X: ⁠⁠⁠⁠https://x.com/anshuman_bh

* Website: ⁠⁠⁠⁠https://anshumanbhartiya.com/

* ⁠⁠⁠⁠Instagram: ⁠⁠⁠https://www.instagram.com/anshuman.bhartiya

Contacting Sandesh

* LinkedIn: ⁠⁠⁠⁠https://www.linkedin.com/in/anandsandesh/

* X: ⁠⁠⁠⁠https://x.com/JubbaOnJeans

* Website: ⁠⁠⁠⁠https://boringappsec.substack.com/