I build production web applications, AI-assisted product workflows, and reliable frontend-backend systems using React, Next.js, TypeScript, FastAPI, PostgreSQL, and Redis.

My strongest work is in end-to-end product engineering: shipping user-facing apps, improving performance, building async AI workflows, and making systems reliable in real usage conditions. I also use AI coding tools like Claude, Cursor, and Codex to speed up development, debugging, refactoring, and code review without treating them as a replacement for engineering judgment.

• Full-stack web applications with React, Next.js, TypeScript, FastAPI, and PostgreSQL
• AI-assisted product workflows using OpenAI API, Gemini API, fixed-schema responses, and async processing
• AI-assisted engineering workflows using Claude, Cursor, and Codex for development, debugging, refactoring, and review
• Offline-first systems for low-network environments
• Async backend processing with queues, workers, idempotent jobs, and webhooks
• Frontend performance, SSR/SSG tradeoffs, and production reliability

Role: Software Engineer Focus: EdTech products, AI workflows, offline-first systems, and production applications

• Owned 4 production apps end-to-end as a founding engineer across products used in 117 schools.
• Built async AI evaluation workflows with Redis queues, FastAPI workers, and Google Gemini.
• Helped reduce teacher evaluation time from 40 minutes to 5 minutes.
• Designed idempotent AI job handling to safely process 5,000+ student submissions.
• Built offline-first flows using IndexedDB queues and batched sync for low-network schools.

Note: Professional project source code is private.

Role: Full Stack Developer Focus: Creator platform, video content workflows, frontend performance, and API integrations

• Built platform modules for video management, publishing schedules, and payments.
• Improved page load performance by moving static catalog pages from over-fetched SSR to SSG.
• Reduced image API response time by profiling resizing bottlenecks and improving processing strategy.

Note: Professional project source code is private.

What it is: A mock interview marketplace where interviewees book 1:1 video sessions and receive AI-generated structured feedback after recorded calls. My role: Built the full-stack product architecture and implementation. Tech used: Next.js, React, TypeScript, FastAPI, Python, Supabase PostgreSQL, Clerk, Stream Video, Google Gemini, Docker, GitHub Actions, GCP Cloud Run Why it matters: Shows full-stack system design, role-based auth, transactional booking logic, video workflows, async webhooks, AI feedback generation, and CI/CD deployment. Link: https://github.com/saurav02022/ai-interview

What it is: A resume tailoring platform where users upload a resume, add a target job description, review AI-assisted changes, and export LaTeX/PDF output. My role: Built the end-to-end product across frontend, backend, AI workflow, testing, and deployment. Tech used: Next.js, React, TypeScript, FastAPI, Google Gemini, Playwright, GitHub Actions, Vercel, GCP Cloud Run Why it matters: Shows product thinking, document parsing, schema-driven AI output, multi-step UX, E2E testing, and deployment automation. Link: https://github.com/saurav02022/resume-builder

What it is: A voice-AI-assisted operations dashboard for verifying high-risk COD orders before dispatch using automated outbound calls. My role: Designed and built the full-stack workflow, backend integration, webhook handling, and deployment setup. Tech used: Next.js, TypeScript, FastAPI, Bolna, Firestore, Docker, GitHub Actions, GCP Cloud Run Why it matters: Shows practical full-stack engineering around real operational workflows, third-party voice AI integration, webhook idempotency, polling fallback, and cloud deployment. Link: https://github.com/saurav02022/bolna-assignment-web-app

Frontend: React.js, Next.js, TypeScript, JavaScript, Tailwind CSS, shadcn/ui Backend: FastAPI, Node.js, REST APIs, PostgreSQL, Redis AI / LLM Integrations: OpenAI API, Google Gemini API, DeepSeek API AI Development Workflow: Claude, Cursor, Codex Cloud & DevOps: GCP, Docker, GitHub Actions, Vercel Testing & Monitoring: Playwright, Cypress, Sentry, PostHog Tools: Git, JIRA

• Building reliable full-stack products for real users
• Improving frontend performance and production stability
• Designing AI workflows that are predictable, testable, and safe to retry
• Using AI coding tools to move faster while keeping code quality, architecture, and ownership clear
• Strengthening system design, backend reliability, and product engineering depth

• LinkedIn: https://linkedin.com/in/saurav02022
• GitHub: https://github.com/saurav02022
• Portfolio: https://saurav02022-portfolio.vercel.app
• Email: sk729584@gmail.com