Much of my work at Instagram can be summed up in three themes: Advocate, Delight, and Volunteer.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎ SEPT ‘22—SEPT ‘25. This post is part reflection, part advice (for whoever ends up in this corner of the internet), and a look at some things I built along the way. Rest is either waiting to launch, or living in the Instagram vault forever.

Advocate

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ My product thinking leveled up at Instagram thanks to the rare setup of working on a small team (Studio) pursuing novel 0–1 ideas with the potential to meaningfully change how friends connect on Instagram. I was lucky to have a team that made space to engage deeply with each other’s ideas through weekly product discussions and I got to be in tight, focused working sessions with a PM or an engineer often.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎We did this for all phases of an idea, not just the initial brainstorm. We debated everything from first principles and built a culture around strong opinions (not just welcomed, but required). But those opinions were loosely held, which meant you weren’t advocating from ego, you were advocating for the best idea to win. That made everyone think sharper, listen harder, and push each other and each idea to great lengths.

[SOMETIMES IT MEANT BEATING IDEAS TO DEATH]

[SOMETIMES IT MEANT SHIFTING DIRECTION]

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎Group Mentions (ability to mention multiple people in a single sticker) came out of an idea that was originally centered around events. After many product discussions and a full rework, we landed on something the team had real conviction in. Although at the end, Group Mentions didn’t ship (it just didn’t move the big numbers IG cares about), it’s still the most positively liked and well received Stories feature in Adam Mosseri’s broadcast channel. Wish I could’ve done more to get this one out the door beyond the public test.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎One idea that did ship through a winding, iterative path was Add Yours Music sticker. The project started as an interactive sticker that let users rank their top 3 songs, creators, or opinions. But early prototyping made it clear that it was too much friction. We simplified it to just a one thing, which helped usability but blurred its value. We layered on new mechanics, but it still didn’t feel right.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎In one long conversation in a product discussion, we shifted the direction entirely. We narrowed the focus to music and leaned into sharing your favorite song of the day, and finding out your friends’ too. That led to Add Yours Music↗. It ended up launching as one of the lightweight, expressive sharing options for Stories in our Sticker Drop campaign↗ as the hero sticker w/ Dua Lipa↗.

[SOMETIMES IT MEANT BEATING IDEAS INTO SHAPE]

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎Quicksnap started as one of many camera concepts we explored in our team, and through prototyping and constant dogfooding, it evolved into something special because we were relentlessly advocating for our teen users as we shaped the experience, designing with their creative habits in mind. Quicksnap isn’t fully launched just yet, so for now I’m going to show my work through a few social media posts↗ and screenshots↓ from the public test countries.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎Quicksnap was our attempt to bring back the kind of everyday sharing that felt missing on Instagram. We started by designing a stripped-down camera mode with no filters or edits. Then we built the audience into the shutter, making sharing a single tap — no post-capture. It was quite jarring. To make the idea stick, we had to shape it with new pieces along the way, like a “practice round↗”, a more hands-on onboarding, to help ease anxiety around a one-tap share.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎On the receiving side, Quicksnap was meant to feel like a playful invitation from a friend (a quick glimpse into their day) so I was obsessed with keeping the entry point simple and gestural, always about speed, novelty, and connection. To support that, we pushed audience lists↗ to feel less broadcast-y↗ on the sender side, keeping it focused on friends. And for consumption, media peeked in from the edge of the screen. For me, the peek just added a touch of whimsy that fit the nature of sending and receiving disappearing media.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎In the end, after countless rounds of pushing, refining, and reshaping, the final product brought together all the pieces↗ that made the feature work: quick, low-pressure sharing mechanics, lightweight consumption patterns, a simple interaction model, and setting specific audiences to share with. We also added recap and archive features, and layered in dozens of small touches like animations, haptics, and the occasional easter egg.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎Every part of Quicksnap was pushed into existence by design work, team trust, and a whole lot of advocating WHEN IT WOULD’VE BEEN EASIER TO MOVE ON. That persistence unlocked a whole new way to share daily moments that teens loved, sparking significant organic growth, more overall sharing (not just mix-shift!) and a wave of new casual producers.

[AND SOMETIMES IT MEANT PUSHING PAST NO’S]

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎Keeping ideas alive takes work. You have to advocate in so many different forums, to so many different people, especially when leadership isn’t convinced or is looking for clear numbers to back things up. I have plenty of stories about fighting to keep an idea moving, and a few where I wish I had, but I’ll spare those for now. One of my favorite things I added to the app came from pushing hard for a TINY FEATURE with no data story: making @mention reshares full-bleed.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎Adding a single line of attribution in consumption is not headline-worthy, but it cleaned up a cluttered UI. Though without a metric win, this design stalled in testing, and so I rallied the team to pull user sentiment (incl. screenshots from Reddit and X), which we then reframed it as a craft win to get it shipped. That one small release↓ led to more craft-focused product improvements across Instagram and eventually helped inspire and launch full-screen Reels reshares later that year.

Delight

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎I’ve seen firsthand how much users love surprise and discovery, and it’s one of my favorite things to sneak into every feature I work on. Just enough magic to make people smile or pause. More on those projects, like this mysterious ball↗, when they launch (soon).

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ One of my favorite delight features, Emoji Pong, started as a silly idea I prototyped and posted in the biggest channel I could find, just to see if it got others excited. It did. Within a week, we had a working iOS prototype of the game (triggered by tapping on any emoji in DMs). One more week later, it was live on Android too. The joy it sparked internally made it easy to rally engineers who wanted to build it. We shipped it quietly with no marketing, just to see if it might spark anything. It went viral↗.

[TAKE SILLINESS SERIOUSLY]

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎Overall these types of delight features become brand sentiment for the app. In the case of Emoji Pong, it blew up on TT and Reddit with people breaking down game plays↗, got highlighted in press events↗, and continues to go viral whenever someone finds a new trick↗. I prototyped and shared many other silly experiments that never shipped, but honestly, that wasn’t always the point. Sometimes you make a thing just to get people excited, open up a new direction, or remind the team that play has a place in the product too.

Volunteer

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎During my ~3 years at Instagram, I jumped into nearly 20(!!) design sprints across Instagram and Threads. I also gave over a dozen lightning talks covering everything from product lessons to team insights to a few offbeat topics (for onboarding, IG Design Summit, or anywhere it helped to hear directly from someone deep in the work). I jumped into these moments because I wanted to, not because I had to.

‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎‎ ‎ ‎The design sprints especially reminded me I’m good at rapid idea generation and some of those ideas even made it onto other teams’ roadmaps, shaping real product. I also jumped in to help build culturally relevant, limited-time Instagram features. These were often fast, scrappy efforts run by a small crew in need of help from product designers and engineers. Through them, I’ve gotten a handful of fun, high-energy ideas out into the world, like this ‘Back to School’ activation↗

The Shadow Side

[FEW HONEST FOOTNOTES ON THE THEMES ABOVE]

#Advocate
Because our team was so collaborative and not only the PM but the engineers and the rest of the XFN were deeply involved in product thinking, I had to remind myself that some decisions are design decisions, and not design by committee.

#Delight
Emoji Pong’s success spun up a whole games initiative on Instagram, but the magic wasn’t the game itself, it was that it came from play. It’s easy to try and copy the outcome instead of cultivating the curiosity and freedom that led to it.

#Volunteer
Jumping into everything can stretch you thin, especially when you're raising your hand too often while still juggling your actual work. Every ‘yes’ to a new ask that comes your way costs you something. Make sure it’s worth it.

Wanna know what it’s like to design at Instagram? I gave it a shot trying to illustrate it at the bottom of this page↗