GitHub - rabbilyasar/rabbilyasar · GitHub
Skip to content

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Latest commit

 

History

7 Commits

Folders and files

Repository files navigation

Rabbil Yasar Sajal

Software Engineer · Product Builder · Python

I build backend systems, developer tools, and AI-powered products that solve practical problems.

Portfolio · LinkedIn · Dev.to · Email


Building

EnvShield A developer tool for managing environment configuration as a version-controlled contract, with schema validation, secret scanning, and Git enforcement.

JossJobs An AI-powered job intelligence system that turns noisy job postings into structured, evidence-backed opportunities. The pipeline handles discovery, extraction, eligibility assessment, fit scoring, and ranking while minimizing unnecessary LLM usage.

KemonChilo A restaurant customer intelligence and service recovery platform. Customers provide feedback through QR/NFC, while restaurants get real-time alerts, AI-powered issue classification, operational insights, and structured service recovery workflows.

Agentic Engineering

I'm interested in using AI agents as engineering tools, not just chat interfaces.

I've been experimenting with:

  • Claude Code and agent-driven software development
  • Playwright-based browser automation and testing
  • MCP-based agent tooling
  • Agentic UI testing and verification
  • Benchmarking agent workflows and tool usage
  • Designing workflows that keep deterministic logic outside the LLM
  • Using AI where it provides leverage without turning the entire system into an LLM dependency

Engineering

I mostly work with:

Python Flask Django PostgreSQL Docker AWS Terraform Pytest Playwright

I'm particularly interested in:

  • Backend architecture and APIs
  • Developer tooling
  • AI-assisted software development
  • Agentic systems and automation
  • Testing and developer experience
  • Turning small ideas into production software

How I build

Solve the actual problem. Keep the system understandable. Automate what should be automated. Use AI where it provides leverage. Don't add complexity without a reason.


Building useful things, one problem at a time.

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors