Skip to content
View faisalsardi-dev's full-sized avatar
🎯
Focusing
🎯
Focusing

    Highlights

    • Pro

    Block or report faisalsardi-dev

    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.

    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
    faisalsardi-dev/README.md

    Agentic Development

    I enjoy building software at the boundary between deterministic systems and AI-driven systems, and understanding complex technologies by rebuilding simplified versions of them from first principles.

    My interests include agent workflows, retrieval systems, backend engineering, applied AI, system design, deployment, and infrastructure.

    Many of my projects explore questions such as:

    • When should a problem be solved with code?
    • When should it be solved with an agent?
    • How should both approaches interact safely?
    • What architectural tradeoffs emerge when AI becomes part of the software stack?

    Technical Interests

    Python
    FastAPI
    Go
    javascript
    Agent Systems
    RAG
    LLMs
    LangChain
    SQLite
    non-relational databases
    Linux
    System Design
    cloud security
    aws
    Backend Development
    APIs
    Git
    

    Current Featured Projects

    🚀 Kiosk Agent Order Maker — Enhanced Performance

    A research-driven evolution of the original OrderBuilder system.

    This project compares Prompting, RAG, and Fine-Tuning approaches for converting natural-language food orders into structured order JSON. It includes synthetic dataset generation, a reproducible evaluation harness, deployment infrastructure, and a fine-tuned Qwen3-8B model.

    Repository: kioskagentordermaker-enhanced

    🍔 OrderBuilder

    An AI-powered ordering platform that allows customers to build orders either through a traditional kiosk interface or through a conversational agent.

    Repository: kioskagentordermaker

    💬 Meowic

    A command-line WhatsApp client built on the whatsmeow Go library, exposing multi-device messaging to coding agents either directly through the CLI or through a stdio/HTTP server that wraps the program in the Model Context Protocol that tracks every call the LLM makes.

    Repository: meowic

    ⛓️ OpenLedger

    A simplified cryptocurrency implementation demonstrating wallet generation, transaction signing, balance tracking, and blockchain-style ledger verification.

    Repository: openledger

    🪙 CLICrypto

    A lightweight command-line cryptocurrency tracker that retrieves live market data directly from CoinGecko.

    Repository: clicrypto

    🤖 Leap-Aware Agent

    A comparison of deterministic software architecture and LLM tool-calling workflows using leap-year-aware date calculations.

    Repository: leapawareagent

    Popular repositories Loading

    1. kioskagentordermaker kioskagentordermakerPublic

      Fast food self service kiosk: 1-Traditional order assembly method+2-NLP AI order assembler method

      Python

    2. leapawareagent leapawareagentPublic

      a comparison of an llm-tool system solution vs classic dynamic programming at calculating the number of days between two dates

      Python

    3. openledger openledgerPublic

      A fun simulation of how cryptocurrencies work: your keys never leave local, you broadcast the hashed transaction to the backend which confirms and appends. DEMO VID AVAILABLE

      Python

    4. clicrypto clicryptoPublic

      A fast CLI for live crypto prices, market rank, and dominance — with efficient coin name resolution via the CoinGecko API.

      Python

    5. faisalsardi-dev faisalsardi-devPublic

    6. kioskagentordermaker-enhanced kioskagentordermaker-enhancedPublic

      a fine tuned approach with the menu and ingredient instruction baked into the weights, an approach with Langchain structured output and the whole menu and ingredient instructions thrown into the sy…

      Python