Bio
AI engineer and researcher. Co-presents with Kiriti Badam on AI product design — particularly on the irreducible non-determinism of LLM-backed products and the operating frameworks (CCCD, behavioral flywheels, transparent uncertainty) needed to build them well.
Operating themes
- Non-determinism is the irreducible AI tax. Classical product loops break.
- Continuous Calibration, Continuous Development. The replacement operating loop.
- Transparency in uncertainty. Confidence scores beat opaque authority.
- Behavioral flywheels can go negative. Seed positively, watch the loop.
Cards
- Continuous Calibration, Continuous Development (CCCD) is the operating loop for AI products — CCCD is the operating loop for AI products [Tier A]
- Show model uncertainty in the UI; opaque confidence destroys trust — Show model uncertainty in the UI [Tier B]
Sources captured
- 2026-04-28 — Lenny's Podcast, "AI product design, non-determinism, CCCD" (
raw/podcasts/aishwarya-naresh-reganti-kiriti-badam--ai-product-design-non-determinism--2026-04-28.md)