Skip to content
View ss-sevesh's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report ss-sevesh

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ss-sevesh/README.md

Sevesh S S β€” AI and Machine Learning student at KEC. Building AI that you can use.

AI & Machine Learning student @ KEC Β· Full-stack developer
Exploring local AI, computer vision, and tools that turn information into something useful.

Selected work Β  / Β  Watch a demo Β  / Β  All repositories

From an idea to something you can try.

I like connecting models to real interfaces: a voice you can interrupt, a rally you can replay, or an answer you can trace back to its source. My projects span speech, vision, retrieval, and the full-stack applications around them.

Three areas of my work: voice β€” listen, reason, speak; vision β€” observe, track, explain; knowledge β€” retrieve, ground, connect.

Selected work

Local Voice: English and Hindi speech systems

A local voice companion for English, Hindi, and Hinglish. Speak naturally, see the answer stream in, and interrupt the reply with your voice.

Python Β· FastAPI Β· React Β· Whisper Β· Qwen Β· Kokoro

Watch the audible English + Hindi demo β†’

Open the silent workspace preview. Both speakers are audible in the full demo above.

ShuttleSense: computer vision for badminton

Local AI badminton video analysis: rally replay, shuttle tracking, body poses, heatmaps, and coaching reports grounded in the footage.

Python Β· Computer vision Β· Video analysis

ReelToReal: videos to searchable knowledge

Turn saved short videos into an Obsidian-compatible local knowledge vault, with timestamped notes, speech recognition, visual understanding, and hybrid retrieval.

TypeScript Β· Whisper Β· Moondream Β· Qdrant Β· RAG

Building, one contribution at a time.

Real GitHub contribution calendar with a decorative animated scan

The green cells show real contribution levels. The moving line is decorative; this is a periodically refreshed snapshot, not live telemetry.

Watch the contribution snake

Animated snake traversing my GitHub contribution calendar

Generated from my GitHub calendar with Platane/snk. Animations stop when your device requests reduced motion.

Recently on the workbench

Five latest public development events, with snapshot dates

Read the activity log and open its repositories β†’

Public development events refresh every six hours; GitHub's feed can be delayed.

More experiments, more angles.

Project What it explores
Chemistry Tutor PDF retrieval and local answers with source citations
Manufacturing Quality Predictor An ML prediction system with a FastAPI backend and React dashboard
AQI Model Satellite-based air-quality prediction and leave-city-out validation
Student Attendance System A campus attendance and registration application

Keep learning. Keep building. Explore the source.

Explore the source β†—

How this profile is animated

GitHub displays this README because its public repository matches my username. The header, project illustrations, calendar, activity, and footer are SVG images with CSS animations. They respect reduced-motion preferences and keep their text visible without animation. The linked product preview is a GIF captured from the real application.

GitHub Actions refreshes the calendar and public activity using GitHub's APIs. Generated artwork lives on the output branch. Failed updates preserve the last successful snapshot. There are no visitor counters or generated skill scores. See the design notes for references and editing instructions.

Pinned Loading

  1. shuttlesense shuttlesense Public

    Local AI badminton video analysis: rally replay, shuttle tracking, body poses, rally heatmaps and evidence-based coaching reports.

    Python 1