v
victowolf

Yash Hosalli

@victowolf

Software and AIML Developer

India
Inglese
Alcune informazioni sono riportate in lingua inglese.
Chi sono
Hi! I’m Yash, a software developer and AI/ML enthusiast focused on turning ideas into working products. I specialize in AI/ML, LLMs, full-stack development, Flutter, APIs, blockchain/Web3, and automation. I’m a Google developer in the Dart/Flutter ecosystem, where I worked on Flutter DevTools. I’ve also won hackathons, including an ISRO problem statement at Smart India Hackathon 2025. I can help build AI applications, web/mobile apps, APIs, dashboards, MVPs, and intelligent automation. ... Continua a leggere

Competenze

v
victowolf
Yash Hosalli
offline • 
Tempo di risposta medio: 1 ora

Consulta i miei servizi

Integrazioni IA
I will build ai agents, llm applications and intelligent automation

Portfolio

Esperienza lavorativa

Summer

Developer @ Google (Dart/Flutter)

Summer • Full time

May 2026 - Aug 2026 • 3 mos

As part of Google Summer of Code 2026 with the Dart organization, I worked on extending Flutter DevTools to support WebSocket profiling alongside existing HTTP network profiling. The project addressed a gap in developer tooling: modern Flutter and Dart applications rely heavily on WebSockets for real-time communication, but developers previously had limited visibility into WebSocket activity through the DevTools Network panel. I contributed across the complete tooling stack — Dart SDK (dart:io), Dart VM Service, and Flutter DevTools. I implemented lightweight WebSocket timeline instrumentation covering connection, send/receive, ping/pong, closure, and error events, while keeping runtime overhead minimal. I then built the WebSocket profiling infrastructure to track connection-level and frame-level data such as bytes transferred, frames, timestamps, connection state, and errors, and integrated it with VM Service. Finally, I integrated WebSocket traffic directly into the existing DevTools Network panel, including Overview and Frames inspection views, live updates, filtering, and incremental profiling. The implementation reused existing Network infrastructure rather than creating a separate WebSocket system. The project was completed on schedule, with all four milestones successfully merged into the Dart SDK/Flutter DevTools codebases.