Adrian Amigues

ReactJS - Typescript - React Query - TanStack Router - Material UI Twin is a no-code platform for building autonomous AI agents: users describe what they want done in plain English, and Twin sets up the integrations, browser actions, and triggers needed to deliver it. I joined as a senior front-end engineer while the company was transitioning from a simpler conversational-agent product to the current platform. When I joined, the app was Rust top to bottom, with a Rust back end and a Dioxus (Rust-based UI framework) front end. AI assistance helped the early code along, but as complexity grew the limits showed: slow hot reload, awkward debugging, an immature framework, and no JavaScript-style package ecosystem for front-end work yet. The team moved the front end to React, and that's where most of my work happened. - Front-end architecture: After the move to React, I set up the project structure, tooling, and patterns the team would build on. - AI chat conversation UI: The core interaction in Twin is a conversation between the user and the agent-building system. I built the chat surfaces and the patterns that make those conversations clear and easy to refine over time. - Automatic tool-creation flows: Front-end flows that take what the AI proposes for integrations and tools and make it visible, reviewable, and editable for the user.

ReactJS - Typescript - React Query - Redux - Tailwind - Material UI - Storybook - Jest - React Testing Library - Node.js - Express.js Evaneos built a B2B2C travel-planning platform. Travel agents use it as their daily tool to plan and book trips for their end clients. The codebase was a long-lived React app and on of their cornerstone applications, the link between the travel agents and the clients, was outdated and needed to be modernized to support the upcoming features. That was my main focus during my time there. My work there was focused on analyzing the current state of the codebase for that specific application, and kickstarting the modernization process. - Architectural audit: Reviewed the front-end as a whole: what was working as-is, and what would be problematic once the roadmap features needed to be implemented. I delivered a report of what needed to change and in what order, so the in-house team had a clear modernization plan. - Modernization: Based on the audit, the team started the refactor and I worked on it with them. The main task was to replace the old Redux Thunk + Saga code, which had become impossible to maintain, with React Query. The new setup used React Query's powerful caching capabilities to keep the same UX (and improve it in some areas), with less manual data-fetching code to maintain. - Workflow: I worked within the team's normal workflow: weekly sprints, backlog refinement in dev pairs, peer reviews, pair programming.

ReactJS - Typescript - Nx - Ag-grid - Tailwind - Figma - Jest - Testing Library - Node.js Phacet builds AI tools for document workflow automation. I joined as the front-end lead on a small product team, owning the UI for two products and the shared front-end infrastructure underneath them. The core product was table-centric: a spreadsheet where each row is a document and each column is an extraction the AI runs against it. Users compose workflows by defining columns (natural-language questions or structured queries), the AI fills cells from the source documents, and the user reviews and corrects the output. Most of my front-end work lived inside that table and the Saas around it. It featured: - dynamic and customizable columns, ai-drivent cell population, review and correction UI, chat / table integration, and performance at scale. - AI-driven cell population: cells value streamed, with per-cell status (queued / running / done / errored) and live updates as the back end works through a batch. - Review and correction UI: ability to correct cell values and re-run the extraction with a tweaked prompt. - Performance at scale: keeping the table responsive with hundreds of rows × dozens of AI-driven columns all updating asynchronously. Alongside the main table product I built the AI chat application: a standalone chat UI for the company's agents, with streaming responses, and tool-call rendering. I also set up the design system and the monorepo housing both apps plus the shared design source with Nx and storybook. That's what let a small front-end team ship two products in parallel without the UI drifting between them. Day-to-day I worked closely with product and design in a small team, the product iterations were fast paced, with daily production deploys.

ReactJS - Typescript - Ant-design - Ag-grid - Echarts - Tailwind - Figma - Cypress - Jest - Testing Library - Node.js - Docker - AWS - GitLab Advanced frontend with React on a complex web application showing multiple extensive tables made for advanced data visualisation (with Ag-grid paid version). In addition to the display of data, the app contains multiple complex forms to interact with said data and make modifications that can impact all tables. I was part of the team that setup the displaying of charts (with Echarts library) to visualise that table data in a completely customisable way, where the type of chart, the data in it, and on which axis, is all customisable by the user. When I started I made a frontend codebase state analysis, with suggestion of the areas that could easily be improved, and others that would take more work but that would be important in the long run. Part of the work is also setting up complex e2e testing with Cypress that are critical to the users. Most bugs on the application are critical and can lead to faulty data having potentially huge impacts on the clients decision making that the app recommends. Working full remote with full english communication.

ReactJS - Next.js - Nx - Typescript - Graphql - Chakra - React Router - React hook form - Storybook - Figma - React Query - Jest - Testing Library - Chromatic - Datadog synthetics e2e testing - Node.js - Docker My role was frontend expert here, other developers were all fullstack. So in addition to the technical aspects described further down, I coordinated regularly with the design team regarding best practices and the design system. As frontend expert I was responsible for the frontend architecture of the apps and the libraries we used. I provided training and advices for the fullstack developers that were still learning frontend, and I was always reviewing frontend heavy PRs. I set up a frontend chapter bi-monthly meeting to strengthen the frontend culture in the company. I also participated in improving the global processes planning and delivery agile processes and communication between the product, design, and development teams. The technical aspects were the creation of complex frontend applications implementing a new design system with a focus on UX and accessibility: - Multi applications Monorepo handled with Nx - One big SPA with React Router, and multiple other pages/small apps through Next.js, either SSR or SSG, with Incremental Static Regeneration - Strong typescript typing accros all applications, with backend data typed though Graphql Codegen - Design system implemented with Chakra throughout the different applications - Advanced frontend cache handling and optimistic responses for top-notch user experience with React Query and Graphql - Complex multi step forms with React hook forms with a focus on accessibility - Creation of a custom rich text editor with Tiptap

ReactJS - Typescript - Graphql - Apollo - Styled Components - yjs - Storybook - Jest - Webpack - Node.js - Docker - Figma Advanced frontend and frontend data caching work for a complex project board web application with real time everywhere. - Advanced cache handling with apollo client (subscriptions, optimistic responses, update propagation, relay style pagination) - Fully typed application with typescript and GraphqlCodegen for the api types - Implementation of a new design system with multiple themes - Real time updates through apollo subscriptions - Development of a collaborative editor with yjs - Modern React with hooks and a focus on performance and UX

ReactJS - Typescript - React-Query - Formik - Sass - Css Modules - Storybook - Jest - React Testing Library - ReactI18next - CRA - Webpack - Node.js - Docker - Figma Frontend development in an ambitious quality driven startup: - Feature teams working closely with the backend, design, and product departments - Setup the architecture of three applications interacting together - A focus on quality, with top notch UX through advanced cache manipulations with React-Query - Advanced React with heavy usage of hooks - Active participation in feature planning and decisions

ReactJS - Typescript - Flow - GraphQL - Apollo - Styled-Components - Storybook - Jest - React Testing Library - Enzyme - React-Intl - Node.js - Express.js - Webpack - Docker - Figma - tracking - mParticle - SEO - AWS Frontend development in agile feature teams. in addition to feature development here are some things I did or took part in: - Migration of the entire frontend from flow to typescript - Implementation of a new design system - Setup internationalization on the frontend (France, Spain and England) - Setup react-testing-library for integration testing - Regular updates of our dependencies and tools to improve the team's developer experience and reduce technical debt - Onboarding and teaching the junior developers - Code reviews and pair programming - Multiple refactoring of legacy code into clean modern React

Front end development in React.js and Typescript. - React 16.8 with heavy use of hooks and active refactoring, Typescript, Redux Saga, Jest, Css modules. - Agile team with 1 week sprints, regular backlog refinements in pairs of developers. - Full docker environment for local development. - Git, Github, clubhouse, peer reviews, pair programming. Working on a B2B2C application to provide a platform to agents (external business) and clients to plan their travels. The team was actively removing legacy code replacing it with brand new components with hooks and constantly upgrading outdated packages to improve safety and performance. One of the tasks I got was to do a performance audit of the page. This lead me to learn some advanced usage of the chrome dev tools performance profiler as well as the React dev tool profiler. This lead to several optimisations to the overall performance of the website.

Front end development with React. - Agile environment with scrum, kanban, and present through the shift to an OKR driven organization - React 16.8 with hooks, Redux Saga, Prettier - Unit and Integration tests with Jest - Circle CI, Continuous delivery with Jenkins, peer reviews, advanced git usage Working in OKR driven squads composed of front and backend developers, one UI/UX designer, one QA and one PO. The agile process was fine tuned through agile coaches. In addition the modern web development, my day to day tasks involved taking part in UX workshops, and product explorations to define the team's focus on next iteration. During my time there I added Prettier to the project, poced the use of styled-components to replace the old css architecture in place, and participated to the planning of the future GraphQL architecture.

Front end development with React and GraphQL, working in agile teams on the new responsive website. - JavaScript, React 16, Redux and friends, GraphQL, Node. - Unit and integration testing with Chai, Enzyme, Sinon. - Flow typing, advanced git processes, code reviews, continuous delivery with Travis & Jenkins I had the chance to work with very talented people here, and I got to do several things: - Part of the team that redid the payment funnel from scratch with new designs - I setup storybook - I hosted several retrospective agile ceremonies - Participated in multiple technical interviews for potential recruitements

Research oriented Front End Developer position. The main task is researching new algorithms for locating html elements in javascript when user context changes or the page is updated.

During the fall semester my only course was call "IoT project". I worked on it full time, regular office hours, with a team of 11 fellow students to bring to live a system embracing the Internet of Things principles. We connected physical noise and temperature sensors within a network that would communicate with a Mongo database, an Android application, and another network through APIs. The whole network was created from the ground up using alternative protocols inspired from the Information Centric Network ideas, making this a research project.

My task was to create a new application that would show usage statistics of all the other apps in the suite. It would be used internally and also sold as a product. I used .NET, MySql, CSS, jQuery and several javascript modules.