Waybill / Portfolio 2026
From / sender

Yahya Lendani

Third-year Computer Science student in Warsaw. I build and ship software across stacks: a live subscription product, a local-first transfer app, clinical machine learning and open-source Claude Code skills.

Origin

Warsaw, Poland / local time

Open to
intern­ships

Six shipments, tracked.

Each one carries an honest status. Star counts and last-scan dates are read live from GitHub where the source is public.

YL-CANCELOS

CancelOS

Track every subscription you pay for, see what renews next, and cancel the ones you do not want. A minimalist, iOS-inspired interface with its own DNS mail routing behind it.

Contents
TypeScript
Delivered to
cancelos.com
Last scan
2026-11-05
Source
Private, not declared
Live
YL-NODESYNC

NodeSync

Local-first file transfer between trusted devices. An offline-first, cross-platform app with custom hashing logic for sync, still a prototype and labelled as one.

Contents
Kotlin Multiplatform, SQLite, Supabase
Delivered to
Cross-platform, prototype stage
Last scan
2026-10-28
Source
Private, not declared
In transit. No public link yet.
Prototype
YL-OPUS-TUNER

Opus Tuner

Cost control for long Claude Code sessions on Windows, macOS and Linux: tiered context ceilings, a blocking handoff gate and effort gears.

Contents
PowerShell
Delivered to
Claude Code users
Last scan
2026-11-03
GitHub stars
14
Open source
YL-MOTIONBRIEF

MotionBrief

A Claude Code skill for motion graphics from a brief the agent cannot drift from: a validated brief.json, a build checked against it before render, HyperFrames or Remotion, beat-grid music.

Contents
JavaScript
Delivered to
Claude Code users
Last scan
2026-11-01
GitHub stars
5
Open source
YL-ORCHESTRATOR

Skills Orchestrator

A Claude skill that picks and combines the right skills for a task without being asked, splits big jobs across agents with their own skills, and learns from what worked.

Contents
Python
Delivered to
Claude Code users
Last scan
2026-11-05
GitHub stars
1
Open source
YL-CLINICAL-ML

Heart Disease Clinical ML

Trustworthy deep learning, graph neural networks, conformal triage and causal recourse for cardiovascular risk and MIMIC-IV mortality prediction.

Contents
Python, Pandas, scikit-learn
Delivered to
Research code, not a clinical tool
Last scan
2026-10-29
GitHub stars
3
Research

Contents declaration

What is in the boxes, by category.

Languages
PythonTypeScriptJavaScriptKotlinPowerShellBash
Frameworks
Kotlin MultiplatformTailwind CSSPandasscikit-learn
Data and backend
SQLiteSupabaseDNS routing
AI tooling
Claude Code skills and hooksMCP serversOllamaOpenRouterVertex AI
Environment
Git and GitHub CLIWindows 11macOS
Sender details

Built in Warsaw, shipped everywhere.

I am a third-year Computer Science student based in Warsaw, Poland. I build across different stacks with a strong focus on software architecture, cross-platform applications and, increasingly, machine learning.

I learn by building: training clinical predictive models, designing local file-sync logic, tuning LLM agents. Long term, I want to lead a major tech company as its CEO.

Based in
Warsaw, Poland
Studying
Computer Science, year three
Building now
CancelOS and NodeSync
Looking for
Internships and collaborators

Sign for delivery.