Your recruiter emails to say you've passed the phone screen and they'd like to schedule your "full loop." If you've never heard the phrase, it sounds like jargon for one more call. It isn't. The full loop is the main event — the round where the actual hiring decision gets made — and it looks and feels nothing like the conversations that preceded it. Here's what the term means, what a loop day generally contains, what happens to the feedback afterwards, and how to arrive able to perform in the last session as well as the first.
What "full loop" means
A full loop — also called the onsite loop, the onsite, or just "the loop" — is the final stage of a software engineering hiring process. It's a set of back-to-back interviews with different people, typically four to six sessions of 45 to 60 minutes each, run either as one long day or split across two shorter days. "Onsite" is now largely historical: many loops are conducted remotely, but the name stuck.
The word loop comes from the hiring side. You are handed from interviewer to interviewer around a planned circuit, and each one is assigned a specific competency to probe so that the panel collectively covers the whole job rather than four people asking about arrays.
Where the loop sits in the process
| Stage | Roughly what it is | What it decides |
|---|---|---|
| Recruiter screen | 20–30 min call | Basic fit, level, motivation, logistics |
| Technical phone screen | 45–60 min, one problem | Whether to invest a full loop in you |
| Full loop | 4–6 sessions, 1–2 days | Hire / no hire, and often the level |
| Debrief & review | Days to weeks | Final decision, sometimes team match |
| Offer | Recruiter call | Compensation and start date |
What the sessions usually are
Coding (usually two sessions)
Data structures and algorithms in a shared editor, one problem or a problem with follow-ups, with the interviewer watching you work. What's being assessed is not only whether you land the optimal solution but how you clarify the question, choose an approach, handle edge cases, and test what you wrote. If you haven't sat in a browser-based pad before, our guide to doing well in a CoderPad interview covers the mechanics of that environment specifically.
System design (usually one, sometimes two)
An open-ended architecture prompt — design a URL shortener, a news feed, a rate limiter. There's no single right answer; the interviewer is looking for how you scope requirements, estimate scale, choose components, and reason about the trade-offs of each choice. Junior loops sometimes skip this; senior and staff loops often weight it most heavily. See our system design interview guide for a repeatable structure.
Behavioural or values (usually one or two)
Structured questions about your past work: conflicts, failures, ambiguous projects, times you influenced without authority. Answers are typically expected in a situation-task-action-result shape with specific detail, and interviewers will drill into your individual contribution rather than the team's. Our behavioural interview help page covers building stories that survive follow-ups.
Hiring manager and other sessions
Many loops include a conversation with the manager you'd report to, covering the team's work, your interests, and how you operate. Some companies also add a session run by a senior interviewer from outside the hiring team, whose job is to keep the bar consistent across the company — Amazon's publicly documented "Bar Raiser" is the best-known example of that role. Depending on the company you may also see a domain-specific session on your specialty, or a short values-focused conversation.
How the loop differs from the phone screen
Three differences matter for how you prepare.
- Breadth. A phone screen tests one thing. A loop covers coding, design and collaboration, so a single strong dimension no longer carries you.
- Independence. Multiple interviewers each write their own assessment, generally before comparing notes. One weak session is not automatically fatal; a pattern across sessions is.
- Stamina. A phone screen is an hour. A loop asks you to be sharp for four or five hours, and the last session counts as much as the first.
What happens after: debriefs and committees
The mechanics are worth understanding because they explain the silence that follows.
Immediately after each session, the interviewer writes up detailed notes — usually including what you were asked, what you said, and a recommendation. Most companies deliberately ask for this before interviewers talk to each other, so opinions aren't anchored by whoever speaks first.
Then comes the debrief: the panel and the hiring manager review the written feedback together, discuss disagreements, and reach a recommendation. Interviewers argue in these meetings, and a well-evidenced positive write-up can outweigh a vaguer negative one.
At larger companies there's often an additional layer — a hiring committee of people who did not meet you, deciding from the written packet alone. That's precisely why how you explained your reasoning matters so much: your interviewer can only write down what you said out loud, and the committee only ever sees the write-up.
Everything the decision-makers know about you passes through someone else's notes. Reasoning you kept in your head never makes it into the room.
How long decisions take
One to two weeks is the common range. Smaller companies can move in days; larger ones take longer where committee review, level calibration or team matching is involved. Delays are usually calendar problems, not verdicts — getting five busy engineers into one debrief is the standard bottleneck. Ask at the end of the loop when you should expect to hear, and follow up politely once that date passes.
Preparing for a full day
Preparation for a loop is different from preparation for a single interview because the constraint is energy as much as knowledge.
- Get the agenda. Ask your recruiter for the session list, the format of each, and the platform being used. This is a routine request.
- Prepare per session type, not in general. Coding patterns, a design framework, and six to eight rehearsed stories are three separate workstreams. The 2026 FAANG interview prep guide lays out the schedule, and the FAANG interview prep hub collects the practice material.
- Practise consecutively. At least once, do two or three mock sessions back to back. Discovering how you sound in hour four is the entire point.
- Prepare one strong question per interviewer, and vary them — you'll be asked five times.
- Know the company pattern. If the employer publishes its process, read it. Our breakdown of Microsoft's interview loops is an example of what that looks like in practice.
How loops shift with level
The single biggest source of confusion about loops is that people compare notes across levels without realising it. A new-grad loop leans hard on algorithms and fundamentals, because there is limited work history to examine. A mid-level loop balances coding with a first serious design conversation and behavioural questions about ownership and collaboration. Senior and staff loops invert the ratio: less time on data structures, far more on architecture, technical judgement, cross-team influence and how you handled decisions that turned out badly.
That matters practically, because it tells you where to spend your last two weeks. If you are interviewing above your current title, the marginal hour is almost always better spent on design and on sharpening your stories than on another set of tree problems. If you are unsure which level you are being considered at, ask — recruiters answer this question routinely, and the answer reshapes your whole preparation plan.
Managing the day itself
- Eat properly and keep water within reach. Blood sugar is a real variable across five hours.
- Use the gaps for nothing. Stand up, look away from the screen, breathe. Do not cram between sessions.
- Reset after a bad round. Each interviewer starts fresh; carrying round two into round three converts one weak signal into three.
- Test your kit the day before — camera, microphone, browser, the interview platform link.
- Take notes between sessions. Names and details you'll want for the thank-you note or the offer conversation.
- Book the day off. Trying to squeeze a loop between work meetings is how good candidates arrive depleted.
The short version
A full loop is four to six back-to-back interviews spanning coding, design and behavioural ground, feeding independent written feedback into a debrief and often a committee, with a decision typically landing within one to two weeks. Prepare for each session type separately, rehearse consecutively so stamina isn't a surprise, say your reasoning out loud so it can be written down — and confirm your actual agenda with your recruiter, because every company runs this differently.
Rehearse the whole loop, not one round
Practise coding narration, design structure and behavioural stories in back-to-back sessions with the free-forever plan — no trial timer, no credit card. On the day itself, follow whatever rules your employer sets.
Start Free →FAQ
What is a full loop interview?
A full loop, also called an onsite loop or simply the onsite, is the final stage of a software engineering hiring process: a series of back-to-back interviews with different people, usually four to six sessions run in one long day or split across two shorter days. It typically includes coding, system design, and behavioural sessions, and it follows the recruiter screen and the technical phone screen. The exact number and mix of sessions varies by company, team and level, so confirm your specific agenda with your recruiter.
How many interviews are in a full loop?
Four to six sessions is the common range, each usually 45 to 60 minutes. A frequent shape for a mid-level software engineer is two coding rounds, one system design round, and one or two behavioural or values rounds, sometimes with an additional hiring-manager conversation. More senior candidates typically get more design and leadership time and less algorithm time. Some companies run three sessions, others run seven. Ask your recruiter for the agenda rather than assuming.
How is the full loop different from a phone screen?
The phone screen is a single filter, usually one interviewer and one problem, designed to decide whether the company invests a whole day of engineering time in you. The full loop is the evaluation itself: multiple interviewers, multiple skill areas, and independent written feedback that gets compared afterwards. The loop is also broader — it adds design and behavioural depth that a phone screen rarely reaches — and it tests stamina, because you have to perform at a similar level in the fifth conversation as in the first.
What happens after a full loop interview?
Each interviewer writes up their own feedback and a recommendation, generally before discussing it with the others, so that opinions are not anchored to each other. Those write-ups are then reviewed together in a debrief, and at larger companies a separate committee of people who did not interview you may make the final call from the written packet. The recruiter is your single point of contact throughout. Processes differ by company, so treat this as the general pattern rather than a guaranteed sequence.
How long does it take to hear back after a full loop?
Roughly one to two weeks is typical, though it ranges from a couple of days at small fast-moving companies to several weeks where committee review, level calibration or team matching is involved. Delays are usually logistical rather than a signal about your performance — scheduling a debrief across busy calendars is the common bottleneck. Ask your recruiter at the end of the loop when you should expect to hear, and it is entirely reasonable to check in politely once that date passes.