메타 8년차 시니어가 직접 말합니다 —
개발자 자체는 사라지지 않습니다. 형태가 바뀝니다.
2026년 3월 25일, 제 채널의 첫 게스트로 메타 동료이자 친구인 **제이**를 모셨습니다. 4년차 메타, 총 8년차 시니어 개발자. 그가 보고 있는 시점은 다음과 같습니다.
**"솔직히 괜찮지 않습니다."** 그가 2026년에 유튜브를 시작한 이유 자체가 이 불안 때문이라고 말했습니다. 코드를 쓰던 사람이 하루아침에 코드를 안 쓰는 사람이 되는 변화, 그게 너무 빠르게 체감된다는 것입니다.
**팀 안에서 두 부류로 갈립니다.** AI를 잘 쓰는 사람은 *티를 냅니다* — 사내 발표, 내부 교육, 포스트. 20명 팀에서 한두 명이 AI 교육을 맡는 포지셔닝을 잡았고, 그들은 *"AI 시대에 없어서는 안 될 사람"*으로 자리를 굳혔습니다. 반대편은 평가 척도가 잘못 잡혀 있습니다 — AI **사용량 자체**를 척도로 보는 함정에 빠져 있고, 정작 중요한 임팩트 측정 — *얼마나 빨리, 얼마나 많이, 얼마나 돈을 아꼈는가* — 가 아직 정착되지 않았습니다.
**제이의 결론**: 개발자라는 직업은 사라지지 않습니다. AI를 잘 쓰는 개발자와 못 쓰는 개발자로 갈리고, 못 쓰는 쪽은 대체됩니다. 그게 시니어 본인의 시점에서 보는 그림입니다.
이 대화는 **메타 시니어 코호트가 *내부에서* 이 변화를 어떻게 읽고 있는지**를 보여 주는 episodic record입니다. 외부에서 추측한 게 아니라 본인이 직접 말한 내용입니다.Also in EnglishMeta's 8-year senior engineer says it plainly —
the developer role isn't disappearing. Its shape is changing.
On 2026-03-25 I had my first guest on the channel — **Jay**, a Meta colleague and friend, eight-year senior engineer, four of those at Meta. His view from the inside:
**"Honestly, things aren't OK."** The reason he started a YouTube channel in 2026 was that exact anxiety. Someone who writes code becoming someone who *doesn't* write code overnight — the transition is too rapid to ignore.
**Inside teams, two camps emerge.** The AI-leveraging engineers *make themselves visible* — internal talks, training sessions, posts. A 20-person team typically has one or two people who own AI education. They've positioned themselves as *"indispensable in the AI era."* The other camp has the wrong scoreboard — they measure AI **usage** (tokens, calls) rather than the right metric: *how much faster, how much more, how much money saved.*
**Jay's conclusion**: developers don't disappear. They split into AI-leveraging and AI-displaced. The latter gets replaced. That's the view from inside the senior engineer cohort itself.
This is an **episodic record of how Meta seniors *internally* read this shift.** Not an outside guess — direct testimony.