Joe Taylor built LaserOwl to help enterprise engineering leaders keep tabs on their AI agents.

Joe combines strong product instincts with modern UX judgement and frontend engineering. He can put a working interface in front of a team inside a week, and is unusually decisive about which interfaces are actually worth building. Based in the UK, Joe has spent 15 years across startups and enterprise software in San Francisco, New York, and London.

Joe joined Cisco Collaboration in 2014, where he co-founded the Collaboration Innovation Labs and led the development of TeamTV, an internal video platform for distributed teams. He also built the first user interface for Monica, Cisco’s early AI assistant for Spark and Webex. Earlier work included product engineering at Assemblage from 2012.

In 2018, Joe raised more than $800,000 as CEO of a venture-backed meeting intelligence startup. The company launched Touchbase, a lightweight video collaboration product used by more than 4,000 teams during the pandemic, alongside Nemo AI, a meeting-analysis system trained on thousands of hours of recorded conversations. The products worked. The market timing did not. Running that company end-to-end shaped a pragmatic understanding of product, adoption, timing, and technical execution that now underpins LaserOwl.

Since 2023, Joe has worked as a product and frontend consultant with startups and enterprise engineering organisations, most recently at Harness.io between 2024 and 2025. Much of the thinking behind LaserOwl emerged from observing AI-assisted development workflows inside large engineering teams firsthand.

Enterprise engineering teams are building on AI agents at scale. Those agents are making decisions, calling services, and touching codebases in ways that nobody has visibility into. The session ends. The reasoning disappears. The diff is all that remains.

Most organisations cannot tell you what their AI agents decided last week, which approaches they tried and discarded, or what context drove the code that shipped. When a customer, an auditor, or an incident review asks — there is nothing reviewable.

LaserOwl partners with enterprises to surface that visibility. Not as a developer productivity tool — as an organisational awareness layer. Engineering leaders need to understand how AI is operating inside their systems. LaserOwl makes that possible.

The current engagement is the Agent Development Record Audit. Fixed fee — £15k UK, $20k USA. Self-hosted: nothing leaves your infrastructure. Session capture is deployed via Docker into the customer’s environment, hooks connect to the team’s agents, and 2–3 weeks of passive capture run with no involvement from the engineering team.

The deliverable is a written findings report. At project end, there is the option to continue session logging on your infrastructure as an ongoing engagement.

See what the audit surfaces

Direct email is the fastest route: joe@laserowl.io. Joe is also reachable on LinkedIn.