Building a "Luxury" Software Product in the AI Era: Loïc Houssier, CTO at Superhuman Mail

Mar 12, 2026

Loïc Houssier, CTO at Superhuman Mail, builds product-first, high-quality email experiences with generative AI. He discusses AI features like auto-drafts, semantic search, and personalized triage. Short takes cover building evals from the hardest inbox queries, mixing models, privacy-safe testing, and using coding agents and a quarterly quality week to keep a luxury bar.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

ADVICE

Let Users Combine Prompts With Hard Rules

Give users configurable AI controls: combine subjective prompts with strict rules for precision.
Example: custom auto-labels plus hard rules (like sender-based routing) yield practical, personalized triage.

ANECDOTE

Autodrafts Start With Indexing And Grow Agentic

Autodrafts require indexing user emails and optional knowledge bases to generate context-aware replies that sound like the user.
V1 focused on availability and scheduling; V2 will be more agentic with web search and deeper context retrieval.

INSIGHT

Human In The Loop Prevents Reputation Risk

To avoid AI-generated harm, Superhuman enforces a human-in-the-loop rule: AI drafts are suggested but not auto-sent to prevent 'look foolish' moments.
This is a cultural, product-level safety constraint tied to reputation risk.

Get the Snipd Podcast app to discover more snips from this episode

Get the app

Superhuman charges $30/month for email when Gmail is free. That's always forced them to maintain a different quality bar from most products, and it shapes everything about how they build AI features too.

Loïc Houssier is CTO at Superhuman Mail, and one of the most fun and energized engineering leaders I've gotten to work with. In this conversation, he walks us through what quality really means when you're building a "luxury" software product - and how that mindset applies to AI.

We dig into the high-dimensional challenge of building great AI experiences around email, from auto-drafts to semantic search. We talk about how they approach evals when every user's inbox looks completely different, starting from the hardest queries they can find internally. And we get into how Superhuman is adopting coding agents across their engineering team - including their "quality week" practice and why they removed all procurement blockers for AI tools.

Timestamps:

0:00 - Intro: What "luxury" means for a software product

2:12 - Loïc's background and why he joined Superhuman

6:02 - Game design principles in product development (not gamification)

10:14 - AI features at Superhuman: triage, search, and auto-drafts

15:34 - The challenge of building AI features in a high-dimensional space

20:00 - Building evals from the hardest internal queries (the "wood for my coffee table" example)

24:20 - Privacy and how they handle eval data

25:45 - Their mix of models: BERT, fine-tuned, open source, and frontier

28:19 - What quality means when your competition is free

30:10 - Quality week: dedicating the first week of every quarter to bugs and AI workflow improvements

32:44 - How they're adopting coding agents internally

35:30 - Removing all the blockers for AI tools (e.g. 24-hour security approval, unlimited budgets)

38:06 - How Loïc ramped up on AI as a leader

40:15 - Parting advice: choose your vendors wisely, and enjoy this moment as a builder

Links:

Superhuman: https://superhuman.com
Find Loïc on LinkedIn: https://www.linkedin.com/in/loichoussier/