Sergey – React, React Native, Typescript

Cardinal AI is a vehicle appraisal platform for dealerships, comprising a React Native mobile app and a React admin web portal. There are 2 applications - for Android and for iOS, with the features: on-device VIN capture, multi-step vehicle condition assessment flow, AI-generated appraisal verdicts. Web portal has multi-tenant admin dashboard with financial reporting, inventory, VIN scan history, and role-based access across the dealership hierarchy.

• Delivered two production applications single-handedly: a React Native app and a React admin web portal, including App/Play Store approvals.
• Established the engineering baseline on both repos: feature-sliced module architecture with co-located API hooks/types/validation, code, format, and type checks in CI/CD, and in-repo architecture docs.
• Built an on-device VIN capture pipeline using react-native-vision-camera with OCR and multi-format barcode scanning, including manual input fallback.
• Designed a multi-step vehicle condition assessment flow with per-step Zod schemas, multi-category photo capture with retry/error states, and conditional short-circuit branching.
• Integrated an AI appraisal surface rendering model verdicts, rationale, risk flags and color-coded score breakdowns, hidden behind role-gated access.
• Implemented end-to-end push notifications: permission request flow, token registration and backend sync, token cleanup, deep-linking, and a settings toggle.
• Set up pipelines to build the application bundles and submit to the stores, as well as configured OTA updates for JS-only patches of the app.
• Built a multi-tenant admin panel from scratch covering the full dealership hierarchy.
• Implemented layered role-based access control with route-level guards plus component-level gating, separating the internal admin portal from the dealer-facing area.
• Delivered the core features of the dashboard: financial performance reporting, inventory, VIN scan history, and auction digest views.
• Set up the portal's infrastructure and deployment using Terraform, AWS (S3, CloudFront), and GitHub Actions.

A platform that uses AI to create a digital representation of nonfiction authors that can guide, coach, and converse with readers during and after they finish a book. The platform provides a chat-based experience where users can chat with the author’s representation and listen to the answers with the author’s synthesized voice.

• Translated designs of the application from Figma into a responsive React UI using Tailwind CSS and Shadcn/ui;
• Implemented AI chat flow with the text streaming animation and markdown support using Socket.io and React Query;
• Implemented integration with OpenAI API to build a chat flow for users, including backend streaming and conversation artifact generation;
• Built conversation management functionality for users;
• Extended AI chat feature with a product-specific practices chat, with integration of real-time progress tracking and generation of chat artifacts;
• Implemented authentication flows, including sign-in/sign-up, password reset, email verification, invitations, conversation sharing, and role-based access (admin, author, user) to the platform and data;
• Created smooth animations for UI elements and page transitions using Motion and Tailwind CSS;
• Built a cache layer using IndexedDB for the Read Aloud feature based on ElevenLabs;
• Built end-to-end file upload flows for uploading author books and audio recordings, including file validation, background processing, and storage;
• Implemented templated landing page for marketing purposes using Nunjucks and vanilla HTML/CSS/JS.

Self-care community sharing solutions of guided health experiments that help users treat chronic health issues. The mobile application gives an interactive way to find health experiments, participate in them, measure results, and share them with the community.

• Led a team of 5 full-stack engineers through the full delivery cycle, from feature scoping and estimation to production releases;
• Owned architecture decisions and ran code reviews across web and mobile codebases;
• Communicated directly with the client on priorities, scope trade-offs, and delivery timelines;
• Onboarded and mentored new engineers joining the team;
• Developed a solution for treatment plan creation;
• Built a treatment plan review and community voting system — a human review/approval workflow over user-submitted data;
• Integrated activity and fitness data through HealthKit SDK.

NFT wallet application working with Ethereum, Solana, and Terra blockchains. Wallet helps users to exchange NFTs, integrates with OpenSea and GameFi platforms.

• Implemented UI of the NFT marketplace in React Native app;
• Built Firebase endpoints to manage user assets;
• Implemented integration of NFTs sync from Ethereum and Solana chains.

Talent hiring platform for industrial companies. The platform consists of multiple parts, customer-facing front-end which worked with Next.js, GraphQL API and Node.js microservices.

• Implemented a portal for the talent acquisition team to match candidates with employers using Next.js;
• Implemented role-based authorization for employers, talents, and internal teams to access only relevant pages and data;
• Built API business logic for key platform features such as talent onboarding, candidate management, and collaboration with employers;
• Implemented and maintained the database storage to handle relations between all platform entities and to work with thousands of users.

The product makes health assessments and provides expert advice for mental health, stress management, and other aspects of wellbeing. The mobile application serves as a personal assistant that helps build a nutrition plan and obtain healthy lifestyle habits. The web application provides an interface for the internal management team to run business operations.

• Led a small front-end team delivering the web and mobile applications, running code reviews and task breakdown;
• Directed the design and adoption of a shared UI kit used consistently across the web and mobile codebases;
• Oversaw delivery of an internal operations dashboard for the management team, from requirements through release;
• Drove the team's testing strategy, establishing unit and end-to-end coverage for core business logic.

Nowsta is a powerful tool for managing staff in the hospitality and catering domain. The product provides a solution for scheduling staff, tracking worked hours, gathering statistics, and processing payments. The product consists of web and mobile platforms. The web platform is designed for managers, while the employees use the mobile app.

• Led a front-end team, coordinating feature scoping and the release process via Fastlane;
• Directed the design of advanced schedule management flows, including recurring events, worked-hours tracking, and payment processing UI;
• Drove end-to-end test coverage across the team's codebase.

The application provides inventory management for Shopify storefronts. Using the application, users can manage multiple warehouses in different countries, manage order and exchange flows, and generate operational reports for the team. It is designed to handle many API requests, as storefronts have many user activities related to orders and exchange tracking.

• Implemented application UI and business logic
• Set up the Infrastructure and CI pipelines

Service helps to find English language tutors and schedule classes. It provides interactive lesson materials and a Duolingo-like experience for education progress management.

• Implemented interactive lessons
• Developed marketing websites from scratch
• Communicated with clients directly and improved the product regarding their feedback