6 cards
GitHub Contributions
mxdhavgautam
A local product-eval lab that runs synthetic browser users through real app workflows, records where the experience breaks, clusters the pain, and turns the evidence into repair prompts for coding agents.
A Node CLI tool for explaining where a JavaScript or TypeScript dependency is actually used inside a selected project root.
macOS automation, voice activation, LaunchAgents, custom XTTS voice config, OpenAI Codex context workflows, Agent Driven Development
Friday for Codex is a local macOS voice-triggered wake flow for Codex with a 30-minute background refresh pipeline to keep wake-up fast.
TypeScript, React, yt-dlp, ffmpeg.wasm, OPFS, Chrome Extension APIs
I re-architected Video Grabber from a server/VPS model into a strict client-side Chromium MV3 extension using a background worker, offscreen WASM based yt-dlp and ffmpeg runtime, OPFS checkpoints, and resumable jobs.
Technologies: Deep Learning - MobileNetV2, ResNet152V2, DenseNet201, InceptionV3
This project applies deep learning to classify Cardiac MRI scans using MobileNetV2, ResNet152V2, DenseNet201, and InceptionV3 on a dataset of about 63k images and compares how the results changed with the implementation of Attention Learning into their training loop.
Next.js, TypeScript, PostgreSQL, Prisma, NextAuth, Razorpay, Resend, Vitest
I built a full-stack production commerce platform for a small business with a public storefront, admin dashboard, checkout, order management, analytics, role-based auth, and protected APIs.
6 cards
GitHub Contributions
mxdhavgautamA local product-eval lab that runs synthetic browser users through real app workflows, records where the experience breaks, clusters the pain, and turns the evidence into repair prompts for coding agents.
A Node CLI tool for explaining where a JavaScript or TypeScript dependency is actually used inside a selected project root.
macOS automation, voice activation, LaunchAgents, custom XTTS voice config, OpenAI Codex context workflows, Agent Driven Development
Friday for Codex is a local macOS voice-triggered wake flow for Codex with a 30-minute background refresh pipeline to keep wake-up fast.
TypeScript, React, yt-dlp, ffmpeg.wasm, OPFS, Chrome Extension APIs
I re-architected Video Grabber from a server/VPS model into a strict client-side Chromium MV3 extension using a background worker, offscreen WASM based yt-dlp and ffmpeg runtime, OPFS checkpoints, and resumable jobs.
Technologies: Deep Learning - MobileNetV2, ResNet152V2, DenseNet201, InceptionV3
This project applies deep learning to classify Cardiac MRI scans using MobileNetV2, ResNet152V2, DenseNet201, and InceptionV3 on a dataset of about 63k images and compares how the results changed with the implementation of Attention Learning into their training loop.
Next.js, TypeScript, PostgreSQL, Prisma, NextAuth, Razorpay, Resend, Vitest
I built a full-stack production commerce platform for a small business with a public storefront, admin dashboard, checkout, order management, analytics, role-based auth, and protected APIs.
GitHub Contributions
mxdhavgautamGitHub Contributions
mxdhavgautam