Most companies record what they decided. Almost none record why, or what they considered and ruled out.

Ashutosh Garg, co-founder and CEO of Twin Cafe (formerly Viven), is building Digital Twins: AI assistants trained on how a specific person reasons, decides and communicates, with that person kept in the loop. It is his third company, after Bloomreach and Eightfold, and it started with a blood test, a two-day wait to see his doctor, and three hours with ChatGPT.

Host Jonathan W. Buckley of The Artesian Network and Ashutosh get into:

• Why generic AI chases average knowledge, when people really want the experience of someone they trust
• Keeping the human in the loop: review, edits and escalation to the real person
• A two-month pilot that ended with roughly 140,000 employees on the platform
• Augmentation vs. replacement, and why the quality of work is the real gain
• Whether "digital twin" is the right name for the category
• What enterprise knowledge management looks like in five years
• His advice to founders: trust and security are non-negotiable, and don't build around today's technology constraints

Every episode, on every platform: WeBuiltItBecauseWeHadTo.com

Enterprises rolled out AI agents and generative tools faster than they built any way to see what those systems were doing with sensitive data. Gidi Cohen built Bonfy to ask the question the existing security stack skips.

Most security teams reach for DLP or CASB and ask who has access to the agent. Bonfy's framing — what the company calls Contextual Data Enforcement — asks something harder: what sensitive data is actually moving through AI agents, email, SaaS applications, collaboration tools and generative AI systems like ChatGPT, Claude and Microsoft Copilot. Bonfy's described customer base spans financial services, healthcare, insurance and technology, the sectors where unstructured content is hardest to govern.

Gidi and host Jonathan W. Buckley talk about the intrinsic risks agentic AI brings into an organization and what it takes to get ahead of them, why AI doesn't carry the context a human brings to a judgment call, and the gap between a security problem that has a budget line and one that is merely a priority — the hardest part of selling a category buyers haven't funded yet.

Every episode on Apple, Spotify, YouTube, Amazon and more: https://www.WeBuiltItBecauseWeHadTo.com

Bonfy: https://bonfy.ai

Enjoying the show? Follow We Built It Because We Had To wherever you listen and leave a rating. It helps other founders find these stories. For more on how The Artesian Network helps early-stage tech companies get to market and scale, visit The Artesian Network.

Software ships every week. Most penetration tests still run once a year. Rejah Rehim built Beagle Security to close that gap.

Rejah started his career as a penetration tester, writing scripts to speed up his own audits. He saw small and mid-sized companies skipping pentests entirely because the cost was high and nobody was available to do them, so he turned those scripts into an automated product. Beagle Security launched with a lifetime deal that brought in roughly 1,600 to 1,700 early users, and that early usage trained the system. Today it has around 150 recurring subscription customers, and most still find it inbound through SEO and the blog.

Rejah and host Jonathan W. Buckley talk about fitting security testing into the development cycle, why business-logic testing was the hardest part to automate, how the team handles misses and false positives in a trust-heavy sale, and what AI-written code means for the attack surface. His advice for security founders: get customer feedback as early as you can.

Every episode on Apple, Spotify, YouTube, Amazon and more: https://www.WeBuiltItBecauseWeHadTo.com

Connect with Rejah Rehim: https://www.linkedin.com/in/rejahrehim
Beagle Security: https://beaglesecurity.com

Enjoying the show? Follow We Built It Because We Had To wherever you listen and leave a rating. It helps other founders find these stories. For more on how The Artesian Network helps early-stage tech companies get to market and scale, visit The Artesian Network.