Discover
Zero-Shot Learning
Zero-Shot Learning
Author: Dev Tagare, Nancy Wang
Subscribed: 0Played: 0Subscribe
Share
© 2026 - Zero-Shot Learning Podcast, presented by 1Password
Description
Zero-Shot Learning is a podcast for AI builders, hosted by Nancy Wang, CTO at 1Password, and Dev Tagare, Senior Director and Head of Engineering for Gemini Enterprise & Business at Google. Each episode invites developers from industry-leading AI platforms to break down how AI systems are built and deployed.
10 Episodes
Reverse
Despite the wealth of information LLMs draw on, they are not reliably grounded in the physical world. Spatial knowledge is rarely written into agentic development and remains invisible to LLMs. This episode’s guests, Niantic Spatial’s CEO Inhi Cho Suh and Director of Product Management Eugene Chong, are building a grounding layer to fill that gap. Inhi and Eugene join 1Password CTO Nancy Wang and Google Gemini’s Dev Tagare to discuss why spatial understanding is a prerequisite for AI systems operating in the physical world and the grounding layer for world models to make spatial data precise and useful to AI.
In this episode:
Without an accurate grounding layer, world models are guessing at a world they don’t actually know
Spatial data prevents AI hallucinations by providing “ground truth” instead of predictions of what comes next
Precision enables AI behaviors that weren’t previously possible, like choosing safe delivery drop-off locations
Flipping the sim-to-real process leads to faster and more precise deployment at scale
Zero-Shot Learning is a builder-to-builder podcast about how AI systems are designed, deployed, and secured. Subscribe for more.
Go deeper:
Episode companion blog: https://www.1password.com/blog/physical-ai-authority-model
1Password Developer newsletter: https://1password.com/developer-newsletter
Inhi’s LinkedIn: https://www.linkedin.com/in/inhichosuh/
Eugene’s LinkedIn: https://www.linkedin.com/in/e-chong/
Connect with 1Password
1Password.com
1Password Developer newsletter: https://1password.com/developer-newsletter
Build securely with 1Password Developer: https://developer.1password.com/
In 2020, the SolarWinds attack exposed the vulnerability of supply chains worldwide, a problem this episode’s guest, Matt Moore, had already been working to solve for years. In 2021, he co-founded Chainguard to protect supply chains from future breaches. His conversation with 1Password CTO Nancy Wang and Google Gemini’s Dev Tagare covers why the open source trust model is structurally broken and how AI agents have complicated supply chain security, with a live demo demonstrating how standard tools overlook many vulnerabilities.
In this episode:
AI is shrinking the window for patching, as AI-powered attacks accelerate and demand for pre-alpha code grows
Hacks usually target the delivery pipeline rather than the source code
Net-new, AI agent-written code creates more burden for teams to maintain and secure
Most breaches exploit long-lived credentials
Audited least privilege flips the security model from restricting identities to guarding sensitive resources
Zero-Shot Learning is a builder-to-builder podcast about how AI systems are designed, deployed, and secured. Subscribe for more.
Go deeper:
Episode companion blog: https://www.1password.com/blog/software-supply-chain-identity-security
1Password Developer newsletter: https://1password.com/developer-newsletter
LinkedIn: https://www.linkedin.com/in/mattmoor/
Connect with 1Password
1Password.com
1Password Developer newsletter: https://1password.com/developer-newsletter
Build securely with 1Password Developer: https://developer.1password.com/
Go deeper:
Episode companion blog: https://www.1password.com/blog/ai-agents-as-coworkers
1Password Developer newsletter: https://1password.com/developer-newsletter
LinkedIn: https://www.linkedin.com/in/rohannvarma/
Connect with 1Password
1Password.com
1Password Developer newsletter: https://1password.com/developer-newsletter
Build securely with 1Password Developer: https://developer.1password.com/
While AI agents are already capable of writing and shipping code that works, standard operating procedure still requires human oversight and review. Cognition President of New Enterprise Jeff Wang is building a future where that’s no longer the case. In this episode, Jeff joins 1Password CTO Nancy Wang and special guest co-host Richard Liu, Head of API Products at Anthropic, for a conversation on cloud-based agents and how their ability to prove their own work could make human code reviews obsolete.
In this episode:
Agents need a method to prove that their work is correct
Event-driven agents are absorbing the grunt work nobody wants
Playbooks are the most overlooked and underinvested piece of knowledge infrastructure
Agents should run in isolated sandboxes with only the specific access they need
Code reviews will eventually be replaced by proof that the code does what it’s supposed to
Zero-Shot Learning is a builder-to-builder podcast about how AI systems are designed, deployed, and secured. Subscribe for more.
Go deeper:
Episode companion blog: https://www.1password.com/blog/demoting-the-ide
1Password Developer newsletter: https://1password.com/developer-newsletter
LinkedIn: https://www.linkedin.com/in/jefflinwang/
Connect with 1Password
1Password.com
1Password Developer newsletter: https://1password.com/developer-newsletter
Build securely with 1Password Developer: https://developer.1password.com/
When most developers write code, crash recovery logic is at the bottom of their priorities, until crashes and delays force them to incorporate resiliency logic. That’s why Temporal co-founder and CTO Maxim Fateev built a system to turn code into durable execution. In his conversation with 1Password CTO Nancy Wang and VP of Engineering for Developer and AI Jeff Malnick, Maxim explains why the distributed systems thinking that shaped his work is now essential for building reliable AI agents, with a live demo of a durable agent surviving a crash.
In this episode:
Events are the global variables of distributed systems
Durable execution removes the need for specific languages to run workflows
While agent skills are unreliable, hooks are a hard boundary
Once agents need retries, they inherit the complexity of large-scale distributed systems
Sandboxing agents is the only way to contain what they can access
Zero-Shot Learning is a builder-to-builder podcast about how AI systems are designed, deployed, and secured. Subscribe for more.
Go deeper:
Episode companion blog: https://www.1password.com/blog/build-reliable-agents
1Password Developer newsletter: https://1password.com/developer-newsletter
Maxim:
LinkedIn: https://www.linkedin.com/in/fateev/
Connect with 1Password
1Password.com
1Password Developer newsletter: https://1password.com/developer-newsletter
Build securely with 1Password Developer: https://developer.1password.com/








