2025–2026 xAI AI Coding Interview (Building with AI), New Grad Take-Home / Onsite: 3 Experiences
- Company
- xAI
- Reports
- 3 interview reports
- Level
- New Grad
- Stages
- OA → Phone screen → Onsite
For both new grads, the AI part of xAI's process meant building something. One vibe coded a four-hour take-home in about two hours; the other drew a design without AI at the onsite, then spent about 90 minutes building it with AI.
Coverage: 2 new grad software engineers on xAI's Exceptional Software Engineer track, one interviewed in October–December 2025 and one reported in July 2026, plus 1 take-home report whose hiring level isn't stated, kept in a separate supplement · Take-home, HR and technical screens, onsite · Locations not stated
What came up
Both new grad reports have a step where xAI's AI coding interview meant building something with AI tools, but they put it in different places. In report 02 it was the four-hour take-home, which the candidate built by prompting an AI (vibe coding). In report 01 it was the onsite: whiteboard design first, with no AI, then implementation with AI, then a demo.
| Topic | Reports | Most recent report |
|---|---|---|
| Building with AI tools | 2 of 2 | Jul 1, 2026 (report date) |
| Design without AI, build with AI, then demo (onsite) | 1 of 2 | Jul 1, 2026 (report date) |
| System design screen | 1 of 2 | Jul 1, 2026 (report date) |
| Rate limiters in a take-home | 1 of 2 | Oct–Dec 2025 (interview; reported Jan 28, 2026) |
| Resume deep dive and LeetCode-style coding | 1 of 2 | Oct–Dec 2025 (interview; reported Jan 28, 2026) |
These counts cover the 2 new grad reports, and one report can appear in several rows. The supplement report, outside these counts, also vibe coded the take-home.
Questions we found
AI Coding
| Question | Frequency | All last seen dates |
|---|---|---|
| Insight platform backend take-home: an API rate limiter at 100 requests per second plus a consumer rate limiter at 10 requests per second | 1 report | Oct–Dec 2025 (interview period; report date Jan 28, 2026) |
The task built at report 01's onsite wasn't disclosed.
System Design
| Question | Frequency | All last seen dates |
|---|---|---|
| Design a URL shortener | 1 report | Jul 1, 2026 (report date) |
2 detailed candidate timelines, loops, and screens
01: A whiteboard design, then 90 minutes of building it with AI
Report date: July 1, 2026 · Role: Software engineer, new grad (Exceptional Software Engineer) · Result: Rejected at the end of the onsite, with a six-month wait before reapplying
Recruiter stage
- HR had been in contact since February. The candidate said the timing kept shifting around the company's IPO.
Technical screen (30 minutes)
- No coding at all. The whole screen was one system design question: a URL shortener.
Onsite
- About 30 minutes designing on a whiteboard. No AI here.
- About an hour and a half turning that design into working code, this time with AI.
- A closing 30 minutes to wrap up and demo what they had built.
The candidate kept the actual task to themselves because of an NDA. Their read on the new grad process was that it now rests on system design and has no round of writing code by hand. They were told no at the onsite itself.
02: Two rate limiters, done in half the take-home window
Interview period: October–December 2025 · Report date: January 28, 2026 · Role: Software engineer, new grad (Exceptional Software Engineer) · Experience: Bachelor's · Applied: Referral · Result: Rejected after the first round
Take-home (4 hours)
- The take-home matched what others had already described. The candidate chose the insight platform backend option.
- At its core it was two rate limiters: one on the API, allowing 100 requests per second, and one for consumers, allowing 10 requests per second.
- They vibe coded it and finished in about two hours.
HR call (10 minutes)
- A quick call to set up the rest of the process.
First round (1 hour)
- 15 minutes going through their resume.
- 45 minutes on a LeetCode-style coding problem, which the post names only with a joking nickname. The report doesn't mention AI for this round.
They felt they had answered every question, and the process still ended there.
Supplement: hiring level not reported
This candidate didn't say whether they were a new grad, so the report stays outside the counts above. It covers the same four-hour take-home.
03: Vibe coded overnight, passed two weeks later
Report date: January 23, 2026 · Role and level: not stated · Stage: Four-hour take-home · Result: Passed the take-home
- They started the day after the take-home arrived and worked through the night.
- They vibe coded the entire project, and said they didn't think it could be finished any other way.
- About two weeks after submitting, they heard they had passed. They still didn't know what the graders were looking for.
The report doesn't say which project option they picked.
Keep practicing
- xAI coding questions: every xAI practice problem on FastPrep, mostly algorithm and data-structure questions from xAI's coding rounds. Sign-in required.
- 2022–2026 Cursor, SpaceX and xAI New Grad OA / Interviews: 15 Experiences: reports 01 and 02 alongside new grad loops at Cursor and SpaceX.
- xAI interview guide: xAI's interview process and question list.
- xAI hiring insights: hiring data for xAI.
- Experience feed: candidate reports from every company.
AI coding rounds at other companies
- Meta: Meta AI coding interview: two onsite rounds where you finish a hard algorithm inside a mostly written codebase with an AI model, then face stress-test suites.
- Google: Google AI coding interview: an online assessment with HackerRank's built-in assistant, and a live round on a multi-file project where using AI was optional.
- Stripe: Stripe AI coding interview: experienced engineers kept getting the same transaction rule parser, built step by step with a built-in model.
- Stripe interns: Stripe intern AI coding round: ten intern accounts of the same AI-assisted round.
- Rippling: Rippling AI coding interview: HackerRank's built-in AI on Rippling's familiar problems, plus an HR codebase at the onsite.
- DoorDash: DoorDash AI coding interview: AI CodeCraft: build and run several small services live with your own AI tool.
- LinkedIn: LinkedIn AI coding interview: one long, multi-part task in an IDE whose agent edits files directly.
- Shopify: Shopify AI coding interview: pair programming with your own AI on Shopify's well-known question set.
- Datadog: Datadog AI coding interview: bring any AI tool and share your screen; the tasks are easy, and the follow-ups decide the round.
- Capital One: Capital One AI coding interview: new grad Power Day agentic round: fix a bug, then add a feature to a large full-stack app with Claude Code.
- Bloomberg: Bloomberg AI coding interview: intern rounds that start in a small file and move to a large codebase, debugging with AI help.
- Adobe: Adobe AI coding interview: AI showed up as one step in a director round and in a take-home; the rest stayed classic.
- Snowflake: Snowflake AI coding interview: design and build a rate limiter or a retrying task queue while the AI writes the code; one AI Engineer onsite was in person with your own laptop.
- Snap: Snap AI coding interview: a backend-only chat service from an empty Python project with Claude or Codex, or a feature in an existing codebase.
- Ramp: Ramp AI coding interview: build inside a repo Ramp provides; one repo logged every prompt for the interviewer to read.
- Anthropic: Anthropic AI coding interview: review four pull requests, then improve one with Claude Code.
- Apple: Apple AI coding interview: CoderPad with a sidebar agent, alongside Apple's other recent interviews.
More companies with AI rounds on FastPrep:
- Amazon: Amazon interview report: an AI-assisted debugging OA (1 experience).
- Atlassian: Atlassian interview report: project coding and AI tool use (11 experiences, two parts).
- Circle: Circle interview report: three AI rounds: building, debugging and PR review (1 experience).
- Volta: Volta interview report: a deliberately vague data task with a token budget (1 experience).
Across companies: 2026 AI Coding OAs at 15 Companies: 32 Reported Problems and 2025–2026 AI Coding Phone Screens and Onsites at 43 Companies: 76 Reported Problems.
Sources: Anywhere useful interview info can be found.
Comments
0No comments yet — be the first to share what you know.