Endfield Team Building: The Four Jobs Every Squad Covers

Version note. Every operator named on this page comes from our own character database, which is checked against version 1.4 "Homecoming". Version 1.5 was scheduled for 2 September 2026 — two days before this page was written — and we have not re-checked the roster against it. New operators, changed kits and moved tier placements are all possible. Where a character page and this page disagree, trust the character page: it carries its own updated date. Read the framework as durable and the names as a 1.4 snapshot.

Four jobs, not four names

A squad fails because a job is missing, not because a rarity is missing. Four things have to happen in a fight: something deals the damage, something absorbs it, something puts health back, and something controls the pace — a slow, a stagger, a way out. The names that fill those slots change every patch. The four jobs do not.

That is also how our own operator files are written. Look at the role line on any character page and it names a job rather than a power level. Wulfgard is filed as Damage Tank · Squad Sustain. Bremen is SP Battery · Mob Cleanup. Embers is Mobility · Aerial Disrupter. Two of those three sit at B tier, and they still hold jobs nothing else in our database holds.

One thing this page will not pretend to know: every squad recommendation we have published is written as four operators, so four is the shape used below. Whether the game permits other squad sizes in every mode is not something we have checked, and we would rather call it a house convention than dress it up as a rule.

What each class is for, and what we actually have filed

Our database currently holds 15 operators across 7 classes. The distribution is lopsided, and the lopsidedness is the part worth planning around.

Class	Job it covers	Filed in our database	What that means when you plan
Guard	Front-line damage, in contact with the enemy	Chen Qianyu (S), Surtr-Z (S, not released), Pith (A)	Deepest class we have filed. Pith is a 4★ sitting at A, so this slot is fillable without a 6★.
Caster	Ranged area damage, squad buffs	Perlica (S), Vellica (A), Lummis (B)	Three grades, three budgets. Perlica's file explicitly warns against stacking two casters.
Sniper	Ranged single-target damage, mob lockdown	Ines (S), Hoderi (A), Zorah (C)	Hoderi is the budget answer. Zorah's own file says bench her by chapter 4.
Defender	Holding aggro, cutting incoming damage	Wulfgard (A), Kuranta (B)	Two filed. Kuranta's file positions her as the answer when you do not have Wulfgard, and warns off boss content.
Medic	Putting health back on the squad	Endmin (A), Nymeria (A)	Two filed, both A, neither S. Nymeria's file calls her an account requirement rather than a preference.
Vanguard	Skill-point generation, clearing trash	Bremen (B)	One candidate, a B-tier 5★. If your carry needs extra ult rotations, he is what our database has.
Specialist	Mobility, reaching what melee cannot	Embers (B)	One candidate. Her file is blunt: outside stages with fliers, she is benched.

The two thinnest classes are also the two most situational, which is the trap. If you are holding a slot open until you pull a great vanguard or a great specialist, you are waiting on a class our own database lists exactly one candidate for. Build the one you have, or plan the squad so the slot is not load-bearing.

Six elements, one of them nearly empty

Across those same 15 operators, element coverage runs: physical 4, fire 3, lightning 3, ice 2, nature 2, arts 1. Six element types, and a very uneven pile.

Physical is the easy slot to fill and the easy trap to fall into. Four of our filed operators are physical, including two of the three released S-tier entries, so an account that pulls well ends up physical-heavy without deciding to. Our Ines file leans into that on purpose — she and Chen Qianyu are described as stacking their physical shreds — but stacking a damage type is a bet that the stage in front of you does not resist it.

Our files pair elements for two different reasons, and they are worth telling apart. Endmin plus Hoderi is written up as a lightning pairing — same type, stacked on purpose. Perlica plus Chen Qianyu is arts and physical together, because two different amplifiers do not compete. Only the second kind of pairing still works when the stage changes under you.

Arts is the outlier: one filed operator, Perlica, and her own file asks for the arts amplifier to stay exclusive to her while it is running. That is a coverage note, not a ranking. If arts damage is what a fight wants, our database currently offers you one answer.

Build from your roster, not from a banner

The order that works is boring. Fill the missing job first, upgrade the filled ones later.

1. Damage you can actually aim. Whatever your best-invested attacker is. Our Pith entry exists for exactly this reason — a 4★ at A tier that our file says fits anywhere.
2. Something that survives contact. Wulfgard if you have him; Kuranta if you do not, with the caveat her own page attaches about boss fights.
3. Health going back in. Endmin or Nymeria. Our Wulfgard file makes the trade explicit: skip him and you have to invest in Endmin instead, and Endmin does not bring his elemental damage reduction with her.
4. The flex slot, chosen per stage. This is where our published pairings live — Hoderi against mob-heavy stages, Endmin against single bosses, Embers when there are fliers, Bremen when the carry is starved for ult uptime.

If you want a concrete starting point, the earliest team our files name is Chen Qianyu, Hoderi, Wulfgard and a free starter caster, from Hoderi's own entry. It is not the strongest squad on the site. It is the one assembled from what a new account plausibly owns, which is the more useful property. The first-30-days guide covers what to spend on those four.

A high tier grade is not a strong squad

Our tier list is labelled a fan estimate, and our site FAQ says the same thing in plainer words: treat any Endfield tier claim as a fan estimate, not measured data. Here is the sharper version of that caveat, using nothing but our own file.

All four S-tier entries in our database are damage classes. Perlica is a caster, Chen Qianyu is a guard, Ines is a sniper, Surtr-Z is a guard and is not released yet. A squad built by taking the top of our own ranking has no defender and no medic in it — and Surtr-Z's own entry says she cannot be run in low-sustain compositions. The all-S squad contradicts itself on its own page.

Read down a grade and the point holds from the other direction. Wulfgard sits at A, and his file says he is functionally S in any squad that wants to finish endgame content. Nymeria sits at A as a 5★, and her file says she is often more useful than a second 6★ damage dealer. Pith is a 4★ at A. Grades rank operators against operators. A squad is not four operators ranked; it is four jobs covered.

So the practical rule: a B-tier defender in an empty defender slot beats a second S-tier caster in a slot that was already covered. That is not a hedge about tier lists being unreliable. It stays true even if the grades are perfect.

Five squads that look fine and are not

These are the failure shapes that our own operator files keep warning about, collected in one place.

What the squad looks like	Why it stalls	What our pages point at instead
Four 6★s, no medic and no defender	Nothing puts health back and nothing holds the front. Damage does not matter in a fight you do not finish.	Wulfgard is filed as tank and sustain in one slot; Endmin and Nymeria are the two filed medics.
Two casters stacked for more damage	Perlica's file says she wants the arts amplifier exclusive to her while it runs. A second caster competes rather than adds.	Pair her with a different damage type. Her file names Chen Qianyu as the best partner for that reason.
Everything on one element	Convenient until a stage resists it. Physical is the usual culprit because it is the deepest pool we have filed.	Split by intent. Where our files pair two of the same element, they say why — Endmin plus Hoderi is written up as a lightning pairing on purpose.
No vanguard, and the carry never has an ult ready	Bremen is filed as an SP battery. If nothing is generating skill points, burst comps run dry between windows.	Bremen is the only vanguard in our database. His file points him at burst teams with Perlica or Vellica.
Two damage dealers who want the same debuff window	Our Chen Qianyu file warns against pairing her with Vellica outside niche AoE comps because they compete for mark uptime.	Pick the one that matches the fight length. Vellica's file frames her as the long-fight option.

Four of those five are overlaps, not gaps. That is worth sitting with: the more you pull, the more likely your squad problem is two operators fighting over the same job rather than an empty slot. Gaps are obvious. Overlaps look like a strong roster right up until the clear time stops improving. If a swap is not paying off, the combat system guide covers the ordering side — sometimes the squad is fine and the rotation is not.

When the answer is to not rebuild

Most stages do not need an optimal squad. They need a squad that covers the four jobs at a level appropriate to the content, and a lot of the frustration in this game comes from players re-planning a roster that was never the problem.

Rebuild when something actually stops you: a wall you have failed several times, a stage type your current four have no answer for, a new operator who fills a job you have been faking. Do not rebuild because a ranking moved, because a video called your carry mid, or because you pulled a 6★ in a class you already covered. Levelling a fifth operator to sit on the bench costs the same materials as levelling the one you play.

And the honest caveat on all of it: this page names operators from our 1.4 files. The four-job framework should survive 1.5 and whatever follows it. The names might not.

FAQ

What this page is built on

• Our own character database — all 15 operator entries, specifically their class, element, rarity, tier grade, role line and team notes. Every named pairing and every warning above is quoted from those entries rather than gathered elsewhere.
• Our tier list and site FAQ — for the standing statement that Endfield tier claims here are a fan estimate rather than measured data.
• Deliberately left open — the maximum squad size (we use four because our own write-ups do, not because we have confirmed a cap), and anything introduced in version 1.5, which we had not re-checked when this page went up.
• Not included on purpose — skill multipliers, cooldowns and damage figures. Those belong on the individual character pages where they carry their own version date, and repeating them here would only create a second copy to go stale.

This is an independent fan reference, not an official one. If a pairing here contradicts what you are seeing in your own runs, that is worth telling us — corrections are how the character files get better, and they feed straight back into pages like this one.