Work
Case studies covering the problem, the constraints, and how each system was put together.
8 projects shown
Active
CTO & Technical Co-Founder
Platform behind FoundrOS’s Amal Launch Cohort, a 12-week accelerator for Afghan founders: a participant app for applicants, mentors, jury members and guest investors, a separate admin console, and a NestJS + Prisma API with a queue/cron worker. Trilingual (English, Dari, Pashto) with full RTL.
Active
Software Developer
Brand site and storefront for a Kabul IT-training company: course and certification catalog, CMS-driven pages, checkout with bank transfer, cash and gateways in USD and AFN, an agency section, and English, Dari and Pashto with RTL. Deployed with zero-downtime rolling Docker releases behind nginx.
Active
Software Developer
Udemy-style learning platform: Moodle-style question bank with six question types, timed exams including a PMP-format mock exam with sections and breaks, an exam lock that cancels attempts when the student leaves the window, AES-encrypted HLS video with viewer watermarks, and verifiable certificates.
Active
Backend Team Lead
Led backend architecture for Afghanistan's leading job platform — listings, search, application tracking, and ATS integration. Designed and built a semantic Applicant–Job Matcher microservice using gRPC, Sentence-Transformers, and scikit-learn.
Completed
Solo Developer
Vector embedding-based job recommendation engine using cosine similarity for personalized, scalable job matching.
Completed
Solo Developer
Chat backend in Go: direct and group conversations, AES-encrypted messages, read receipts and delivery confirmations, S3-compatible file attachments, and JWT sessions with refresh tokens. Built on Gin, PostgreSQL (GORM) and Redis.
Completed
Solo Developer
The Python prototype of the jobs.af job recommender, now used inside the core jobs.af platform: a FastAPI service that embeds jobs and applicants with a multilingual sentence transformer, searches them with pgvector, and scores matches on skills, experience and detected job domain in about 780 ms for 200 recommendations.
Completed
Solo Developer
A movie recommendation system with two modes: content-based suggestions from genre similarity, and personalised picks from a user’s favourite genres. Data lives in PostgreSQL, run with Docker.
