Fayis Mahmood
Full-stack developer with 3 years of experience building scalable web applications
Hire this person
Send a job offer directly to this candidate
About
Full-stack developer with 3 years of experience building scalable web applications using React, Next.js, and Node.js. Collaborated with UX/UI designers to deliver end-to-end features. Skilled with Figma for design collaboration, and AI tools including Claude Code, Ollama, Hugging Face, and Transformers.js. Strong problem-solving skills with a client-centric approach.
Experience
Software Engineer - Gistr - Remote
(2024-02)
AI-native smart notebook transforming videos, PDFs, and articles into structured notes. 18,000+ signed-up users and 3,000+ Chrome extension users. Owned full feature lifecycle from MVP to production.
- Built an interactive Source Viewer supporting 11+ source types (PDF, YouTube, podcasts, EPUB, web articles) with annotation, highlighting, and clip capture.
- Designed bidirectional linking between source viewer and editor — highlights, clips, and AI answers stay anchored to source location.
- Built a Notion-like collaborative editor on TipTap (ProseMirror-based) with real-time multi-user editing using Y.js CRDTs and Hocuspocus WebSocket backend.
- Implemented Ask AI with streaming responses over Server-Sent Events; supports source-grounded Q&A, chapter generation for PDFs/EPUBs/podcasts, and timestamped highlight extraction from YouTube/podcasts.
- Shipped a Chrome extension (3,000+ users) with WXT + React + TipTap so users take notes without leaving YouTube or other sites.
- Built an export pipeline to PDF, Markdown, and image formats preserving editor components using headless Chrome and wkhtmltopdf.
Software Engineer - Nudgelabs - Remote
(2024-02 - 2024-12)
Built thejudgement.in, a portal making Indian court judgements searchable and readable, plus the studio's marketing site.
- Built thejudgement.in, a portal for browsing court judgements with filtered search, saved-judgements library, and interactive table-of-contents reader.
- Wrote scrapers and cleaning pipelines converting thousands of legacy judgement documents into structured, readable formats.
- Added AI-powered Q&A on judgement contents and automatic highlight extraction for key passages.
- Built the Nudgelabs studio site (nudgelab.co): Framer Motion interactions, HLS-streamed portfolio videos transcoded with FFmpeg, Sanity + GROQ as the headless CMS.
Education
BA - History - University of Calicut (2021)
