Front End Lesson 2 – | Dataplexa
Front End Engineering · Lesson 2

Role of a Front End Engineer

You'll break down what a front end engineer actually does day-to-day at a production SaaS company — the decisions, the ownership, and the skills that define the role beyond "building UIs."

The Calendar Doesn't Lie

A study of engineering teams at mid-size SaaS companies found that senior front end engineers spend less than 40% of their working hours writing new feature code. The rest goes to code reviews, performance investigations, architecture discussions, accessibility audits, tooling maintenance, and cross-team collaboration. If you're optimising to write more code faster, you're optimising for the wrong thing.

This isn't a complaint — it's a description of maturity. The higher the stakes of the product, the more time gets spent on decisions that prevent bad code from shipping rather than writing good code after bad decisions have already been made. A front end engineer's real value isn't throughput. It's judgement.

The PixelForge Frontend Team's Week
Monday: Sprint planning and architecture review for the new canvas collaboration feature. Tuesday–Wednesday: Feature development and PR reviews. Thursday: performance profiling session after a Lighthouse regression in staging. Friday: Cross-team sync with Design Systems about the new token migration, plus a post-mortem on a layout bug that reached production. That week included roughly 14 hours of coding and 26 hours of everything else. Both halves matter equally.

What this means practically: the best front end engineers are communicators as much as coders. They can explain why a proposed feature would cause a 40 KB bundle increase to a product manager who doesn't know what a bundle is. They can push back on a design that would require 200 extra DOM nodes per canvas object. They translate between the product world and the engineering world constantly — and that translation requires deep knowledge of both.

There's a version of this role that's purely reactive: wait for designs, implement them, ship, repeat. That version exists, mostly at agencies or in large corporations with strict swim lanes. But at product companies — startups, scale-ups, anything building its own software — the front end engineer is a product collaborator. They're in the room when product decisions get made, and their technical input shapes what's possible.

The Six Responsibilities That Define the Role

Peel back the job title and you find six distinct responsibilities. Most engineers are strong in two or three and actively developing the others. The senior engineers — the ones who get the hard problems handed to them — hold all six simultaneously. They don't switch between roles; they weave them together across a single working day.

The six responsibilities of a front end engineer
1
Interface Architecture — Deciding how components are structured, composed, and reused. Every component decision compounds: a bad abstraction in month one becomes technical debt in month six.
2
State Ownership — Determining what data lives where, how it flows through the application, and how it stays consistent across concurrent users and async operations.
3
Performance Stewardship — Keeping the application fast as features are added. Setting budgets, measuring regressions, and making the case to slow down feature work when the performance debt gets too high.
4
Cross-Team Collaboration — Working with designers, product managers, backend engineers, and QA to make sure the front end accurately implements intent — and feeding back when that intent is technically impractical.
5
Quality Assurance — Writing tests that actually catch regressions, reviewing other engineers' code carefully enough to catch logic errors, and maintaining standards so the codebase doesn't degrade sprint by sprint.
6
User Empathy — Staying connected to how real people use the product. The best front end engineers catch usability problems during implementation — before they ever reach a user research session.

Notice that only responsibilities 1, 2, and 5 involve writing code directly. The other three are about communication, measurement, and understanding people. That balance is intentional — and it's what distinguishes the role from a general software engineering position where the interface is just another module in the system.

Junior, Mid, Senior: How the Role Evolves

The responsibilities don't change as you grow — the scope of them does. A junior engineer owns their own components. A mid-level engineer owns a feature area. A senior engineer owns the entire front end system. The difference isn't the type of work; it's how much of the system you're responsible for keeping coherent.

Responsibility Junior Mid-level Senior
Interface Architecture Implements components from designs and specs Designs component APIs for a feature area Sets the component model for the entire application
State Ownership Uses existing state patterns correctly Introduces new state slices for features Decides the state architecture and migration strategy
Performance Avoids obvious anti-patterns when told about them Profiles and fixes performance issues independently Sets budgets, monitors production, allocates time to fix regressions
Cross-Team Work Asks questions in refinement, follows specs closely Flags technical concerns during design review Shapes product decisions by surfacing technical constraints early
Quality Writes unit tests for their own code Reviews PRs thoroughly, adds integration tests Owns the testing strategy and code review culture
User Empathy Uses the product regularly, reports bugs they find Joins user research sessions, advocates for UX improvements Connects engineering decisions directly to user outcomes in planning

At PixelForge, the jump from mid-level to senior happened fastest for engineers who stopped asking "what should I build next?" and started asking "what's slowing the team down?" Unblocking others — through better tooling, clearer documentation, or well-timed architecture decisions — multiplies output faster than any individual contribution can.

The Multiplier Mindset
Linear contribution: you write 200 lines of good code, the product gets 200 lines better. Multiplier contribution: you document the canvas component API clearly enough that three other engineers can use it without asking you questions, write a lint rule that prevents the most common performance anti-pattern, and catch a state management mistake in a PR before it merges. That week, your direct coding output might be lower — but the team shipped three times as much correct code. That's the senior mindset.

Where Front End Engineers Sit in the Org

The organisational position of front end engineers varies more than most roles in software. At some companies they're embedded in product squads alongside backend engineers and designers. At others they belong to a dedicated front end platform team. At larger companies like Shopify or Atlassian, they exist in both structures simultaneously — platform engineers providing shared infrastructure, product engineers consuming it.

Embedded in Product Squads
Deep product context, fast feedback from users, tight designer collaboration. Risk: front end standards drift across squads without a central platform team to enforce them. Common at early-stage startups.
Dedicated Front End Team
Strong shared standards, consistent tooling, easier to enforce performance budgets. Risk: can become disconnected from product priorities and end up building infrastructure nobody uses. Common at mid-size companies.
Platform + Embedded Hybrid
Platform team owns the design system, build tooling, and shared libraries. Product-embedded engineers own feature delivery. Shopify, Linear, and Notion run variations of this. Scales well above 50 engineers.
PixelForge's Structure
The Design Systems and Platform teams provide shared infrastructure. The Frontend team owns product features. Engineers rotate between the two tracks annually to prevent siloing and spread systems knowledge.

None of these structures is inherently superior. The right one depends on the company's size, the product's complexity, and the engineering culture. What matters is that every front end engineer understands how their team relates to the broader organisation — because that context determines where they should focus their energy.

The Skills Nobody Puts in the Job Listing

Job listings ask for React, TypeScript, CSS, and testing experience. Those are table stakes. The skills that actually differentiate front end engineers in production environments — the ones that come up in performance reviews and promotion conversations — are almost never listed. They're assumed at senior level, and they're what this course trains you toward from lesson one.

Skills that get you hired

React component patterns and hooks

TypeScript type definitions and generics

CSS layout — flexbox, grid, responsive design

Unit testing with Jest or Vitest

REST API consumption and async patterns

Skills that get you promoted

Reading a flame graph to find render bottlenecks

Communicating technical tradeoffs to non-engineers

Designing a component API other engineers can use intuitively

Estimating the performance cost of a feature before building it

Writing code reviews that teach, not just gatekeep

The left column is learnable from documentation. The right column is learnable only from building things that break in production and understanding exactly why they broke. This course accelerates that second category by engineering the production scenarios into the lessons themselves — so you encounter the hard problems in a controlled environment before you encounter them in a live codebase.

The PixelForge Canvas Incident
In Q3, a PixelForge engineer added a feature that queried the document store on every mouse move event inside the canvas. On small documents it was undetectable. On a 300-object canvas, it produced 600 store reads per second and dropped frame rate from 60fps to 11fps. The engineer knew React. They knew the store API. What they hadn't yet developed was the instinct to ask "how often will this code run?" before writing it. That instinct — performance intuition — is one of the right-column skills. And it's taught in Lesson 14.

The gap between the two columns isn't a skills gap — it's an experience gap. And the most effective way to close an experience gap faster than raw time is to expose yourself to the right problems in the right order. That's the architecture behind this course.

How to Read Every Lesson That Follows

Every lesson from here is written from inside a front end engineer's perspective — not as a reference document, but as a working session. When the PixelForge Performance team needs to cut LCP from 4.1 seconds to under 2.5, you're in the room. When the Frontend team debates whether to use Zustand or Context API for the canvas selection state, you're at the table. The technical material is the same as any course. The frame is different.

Read each lesson asking three questions: What decision is being made here? What would go wrong if this decision were made badly? How would I explain this decision to a teammate? Those three questions are what turn technical knowledge into engineering judgement.

Coming up in this module: Lesson 3 maps the full web architecture that sits behind every PixelForge page load — from DNS resolution to the first painted pixel. Understanding that architecture is what makes every performance and rendering lesson that follows click into place.

Quiz

1. A new hire at PixelForge expects to spend most of their time writing feature code. Based on how the role actually works at a production SaaS company, roughly how much of a senior front end engineer's time goes to writing new code?

2. The PixelForge Frontend team keeps seeing the same performance anti-pattern appear in new PRs. A junior engineer fixes it in their own code. A senior engineer takes a different approach. Which action best demonstrates the multiplier mindset?

3. PixelForge is interviewing candidates for a senior front end role. Which of the following skills is most likely to differentiate a senior candidate from a strong mid-level candidate — even though it's not listed in the job description?

Up Next
Web Architecture Overview
The PixelForge Platform team traces every layer of the stack a page request travels through — from the browser's address bar to the first painted pixel on screen.