Skip to content
View AFR0011's full-sized avatar

Block or report AFR0011

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
AFR0011/README.md

Ali Farrokhnejad

Software Engineer · Backend & Full-Stack Engineering · Applied AI

LinkedIn · ORCID

MSc Computer Engineering. I build reliable software systems across backend/full-stack engineering, local-first applications, applied AI, and research tooling.

I'm especially interested in reliability problems around offline state, synchronization, asynchronous work, retrieval failures, security boundaries, and reproducible evaluation.

Selected engineering work

Self-hostable local-first workspace built with Next.js, TypeScript, PostgreSQL, Prisma, and IndexedDB, with user-scoped persistence, offline mutation queues, idempotent synchronization, recovery mechanics, and end-to-end verification.

Personal finance application focused on local-first product engineering, with IndexedDB persistence, authenticated synchronization boundaries, reporting, data portability, PWA behavior, and automated verification.

Independent English/Turkish retrieval assistant for public university regulations, combining deterministic routing, hybrid retrieval, cited answers, refusal and clarification behavior, evaluation tooling, FastAPI, and optional local-model support.

Backend system for isolated asynchronous jobs with ownership-scoped access, atomic resource admission, bounded concurrency, failure cleanup, path confinement, and automated security and reliability tests.

Research software

Validity-aware evaluation framework for two-stage machine-learning intrusion detection. It tests whether strong benchmark results survive stricter checks on class support, schema assumptions, final-system behavior, held-out attack families, rejection behavior, uncertainty, operating-point selection, and reproducibility.

Core stack

Languages: Python · TypeScript · JavaScript · SQL · PHP
Web & backend: React · Next.js · FastAPI · Flask · Node.js / Express
Data: PostgreSQL · Prisma · MySQL · IndexedDB / Dexie · Supabase · Qdrant
AI / scientific computing: PyTorch · scikit-learn · pandas · NumPy · SciPy
Engineering: Git · GitHub Actions · Docker · pytest · Playwright · Vitest · CI/CD

Current focus

Backend and full-stack systems · local-first software · reliability and failure handling · applied AI and retrieval systems · research software · reproducible evaluation

Pinned Loading

  1. ContextOS ContextOS Public

    Local-first workspace with user-scoped persistence, offline mutation queues, idempotent sync, recovery semantics, and end-to-end verification.

    TypeScript 1

  2. Ravel Ravel Public

    Local-first multi-currency finance ledger with offline capture, explicit sync boundaries, reporting, backup/restore, and PWA support.

    TypeScript 1

  3. SignalForge SignalForge Public

    Backend-focused Flask and Socket.IO app for isolated asynchronous media jobs, bounded resources, session-scoped authorization, and tested failure recovery.

    Python 1

  4. EMUAdvisor EMUAdvisor Public

    Local English/Turkish RAG assistant for university regulations with hybrid retrieval, citations, refusal/clarification behavior, and evaluation tooling.

    Python 1

  5. VERA-IDS VERA-IDS Public

    Validity-aware evaluation framework for machine-learning-based intrusion detection.

    Python 1

  6. DartClub-Platform DartClub-Platform Public

    Legacy PHP/MariaDB club platform with role-aware membership, community publishing, and verified five-format tournament workflows.

    PHP 1