Improving is committed to building a great place to work by cultivating an environment that fosters professional and personal relationships. We value open communication, personal growth, and shared rewards, which result in sustainable success.
Voted “best place to work” numerous times, Improving strives to create and maintain a culture that exemplifies teamwork, excellence, and fun! We believe this kind of culture encourages both the inspiration and the motivation to achieve amazing things.
We're hiring a Senior React Engineer to build the next generation of web applications. All of it runs on a shared React and TypeScript platform.
This team ships with AI agents as a core part of the delivery pipeline, not a side experiment. You'll design, measure, and run multi-step agentic workflows that take work from ticket to tested pull request with minimal manual hand-holding — while keeping quality high and token spend low.
What you'll do
Build and own production features in React and TypeScript across guest-facing and internal applications, from component design through deployment.
Design reusable, strongly typed component libraries, hooks, and API contracts that both people and agents can work with safely.
Write structured prompts, agent skills, and command files that reliably produce production-quality code, tests, and documentation.
Build evaluation suites (for example, Promptfoo configs or scripted test cases) that score agent output against clear pass/fail criteria, and iterate on prompts using measured results.
Wire individual agents into end-to-end workflows — such as ticket plan implement test review PR — with defined handoffs and branching logic.
Put automated guardrails between workflow steps: type checks, lint rules, test gates, CI hooks, and adversarial review agents that challenge earlier output.
Define clear "stop and ask a human" points for decisions that shouldn't be left to an agent, and prove they actually trigger.
Instrument workflows so every output can be traced to the step, model and token cost that produced it; use that data to cut spend and raise end-to-end success rates.
Mentor engineers on TypeScript, React patterns, and disciplined agentic development.
Required qualifications
React and TypeScript
6+ years of professional front-end development, with 4+ years building production React applications.
Deep TypeScript expertise: strict mode, generics, discriminated unions, and typed API layers (e.g., generated from OpenAPI or GraphQL schemas).
Strong command of modern React: hooks, Server Components, Suspense, and state and data-fetching tools such as TanStack Query, Redux Toolkit, or Zustand.
Experience with Next.js or a comparable framework, Vite or similar build tooling, and monorepo setups.
A testing habit: unit and component tests (Vitest/Jest, React Testing Library) and end-to-end tests (Playwright or Cypress) running in CI.
Working knowledge of accessibility (WCAG 2.2 AA), web performance (Core Web Vitals), and responsive design for consumer traffic.
Agentic engineering
Daily, hands-on use of coding agents (e.g., Claude Code, Cursor, GitHub Copilot agent mode) on real production work — not just autocomplete.
Writes structured prompts (role, task, context, constraints, or a similar framework) where every instruction has a clear purpose and nothing is filler.
Has built evaluations for prompts or agents with three or more distinct quality criteria and explicit pass/fail thresholds.
Iterates prompts with evidence: records a baseline, forms a hypothesis, makes one change, and measures the result.
Has connected multiple agents or prompts into a multi-step workflow with defined handoffs, run from a harness such as GitHub Actions, a Makefile, or a custom runner.
Builds deterministic guardrails between agent steps (hooks, schema validation, type/lint/test gates) rather than relying on the model to behave.
Tracks cost: can report tokens in, tokens out, and dollar cost per step, and has reduced token spend through model routing, context trimming, caching, or prompt compression.
Gets strong results from mid-tier models and saves frontier models for the steps that need them.
Nice to have
Experience with Promptfoo, LangSmith, Braintrust, or a similar eval framework.
Built adversarial or reviewer agents that check another agent's output before it moves forward.
Structured logging or tracing for agent runs (OpenTelemetry or similar) that ties each output back to the step that produced it.
Has authored Model Context Protocol (MCP) servers, agent skills, or custom tool integrations.
Design-system work with Storybook and visual regression testing.
Background in hospitality, entertainment, e-commerce, or booking and payments flows.
Has trained or mentored others on AI-assisted development practices.
Improving welcomes people from all backgrounds and walks of life, and this is reflected in our largely diverse community of Improvers. Improving is proud to be an equal opportunity employer and is committed to providing equal opportunities for all employees and applicants. Improving recruits, hires, trains, promotes, compensates, and administers all personnel actions without regard to race, color, religion, sex, sex stereotyping, pregnancy (which includes pregnancy, childbirth, and medical conditions related to pregnancy, childbirth, or breastfeeding), gender, gender identity, gender expression, national origin, age, mental or physical disability, ancestry, medical condition, marital status, military or veteran status, citizenship status, sexual orientation, genetic information, or any other status protected by applicable law.
About Improving
Improving is a modern digital services company dedicated to positively changing the perception of the IT professional. We offer innovative solutions through consulting, software development, and training to help thousands of our clients achieve new heights in a competitive and ever-changing market.
As our company continues to grow, we are looking for enthusiastic thought leaders to join our team. Improving has a unique mix of passionate professionals who strive to grow and thrive in new ways. We are committed to establishing and maintaining an inclusive culture that allows all Improvers to bring their authentic selves to work each day. This is why we work hard to build inclusion and diversity in our workplace, so we can all do amazing things and succeed together.
#J-18808-Ljbffr