SE Radio 715: Sahaj Garg on Designing for Ambiguity in Human Input
From Software Engineering Radio - The Podcast for Professional Software Developers by SE-Radio Team
April 8, 2026 · 48 min · Episode 715
About this episode
Sahaj Garg discusses designing AI systems to handle the ambiguity in human input.
Sahaj Garg, co-founder and CTO of Wispr, a voice-to-text AI that turns speech into polished writing, talks with host Amey Ambade about designing systems for the ambiguity that's inherent in human input (text, voice, multimodal). Sahaj focuses on concrete architectural and training strategies for building robust AI systems. This episode examines the problem of ambiguity, where it shows up, building robust systems, personalization, communicating uncertainty, and evaluation. The conversation starts by exploring the difference between inherent and reducible ambiguity, major categories of ambiguity including lexical, syntactic, and pragmatic, and the additional sources of ambiguity in voice, such as homophones and accents. Garg details how to build systems through model training, including providing additional context and constructing datasets for good annotation. They discuss personalization with a focus on "revealed preferences"—learning from user behavior without explicit feedback—and fighting the problem of AI writing that "regresses to the mean." Finally, they consider how to communicate uncertainty to users without degrading the experience, as well as methods for evaluating…
People in this episode
Host: Amey Ambade
Guest: Sahaj Garg
Topics covered
- ambiguity in human input
- AI systems design
- personalization
- model training
- communication of uncertainty
- evaluation of ambiguity resolution
Keywords
- voice-to-text
- AI
- ambiguity
- personalization
- model training
- user behavior
- evaluation
Mentioned in this episode
Organizations: Wispr
More episodes of Software Engineering Radio - The Podcast for Professional Software Developers
- SE Radio 723: Dave Airlie on Linux Kernel Maintenance · June 3, 2026 · 1h 9m
- SE Radio 722: Dwayne McDaniel on the Engineering Challenges of Secrets Management · May 27, 2026 · 52 min
- SE Radio 721: Rob Moffat on Risk-First Software Development · May 20, 2026 · 53 min
- SE Radio 720: Martin Dilger on Understanding Eventsourcing · May 13, 2026 · 56 min
- SE Radio 719: Birol Yildiz on Building an Agentic AI SRE · May 6, 2026 · 54 min
- SE Radio 718: Will Sentance on JS Modernization · April 29, 2026 · 59 min
Explore listener stats, chart rankings, contacts and more on the Software Engineering Radio - The Podcast for Professional Software Developers podcast page.