Skip to content
View SM649's full-sized avatar

Block or report SM649

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
SM649/README.md

header

Typing SVG

LinkedIn Portfolio Email Profile views

🧠 About Me

I'm a Software Engineering graduate from the National University of Modern Languages (NUML), driven by a deep curiosity about how machines learn, reason, and make sense of the world. I work across AI/ML, web development, and software quality engineering, and I'm especially interested in what surrounds the model β€” the reliability, infrastructure, and architecture that make AI systems actually usable in the real world.

  • πŸ”­ I work on AI-driven applications, NLP pipelines, and reliable web systems
  • 🌱 Currently going deeper into RAG, prompt & context engineering, and system architecture
  • πŸ§ͺ Background in SQA & automation testing (Playwright, CI/CD with GitHub Actions)
  • πŸ“ Published research: Beyond the Model: The Missing Layer in AI-Driven Systems
  • 🀝 Part of the @Orenda-Project organization
  • πŸ’¬ Ask me about NLP, sentiment analysis, Flask apps, or test automation
  • πŸ“« Reach me at smraja649@gmail.com

πŸ› οΈ Tech Stack & Tools

Languages

Frameworks & Libraries

AI / ML

Databases

AI Tools & Coding Assistants

Tools & Workflow


πŸš€ Featured Projects

Project Description Tech
πŸ“± Yours A personal AI mobile app that runs open-source LLMs (Qwen3, Gemma) directly on-device β€” private, free, no cloud dependency. TypeScript Β· GGUF Β· On-device AI
πŸ€– Local GPT with Ollama A self-hosted LLM interface that runs AI entirely on local hardware β€” no cloud, no data leaks, full control. Python Β· Ollama Β· JavaScript
πŸ” AI Security Research Research and tooling exploring vulnerabilities, attack surfaces, and defensive strategies in AI-driven systems. Python Β· JavaScript
🧠 Qwen3 QA Fine-Tune Fine-tuning the Qwen3 LLM on a custom Q&A dataset for domain-specific inference and improved response quality. Python · Jupyter · Qwen3 · HuggingFace
🎬 Intelligent YouTube Comment Analyzer Real-time YouTube comment analysis surfacing sentiment trends, topics, and emoji feedback with auto-generated summaries. Final Year Project. Flask · BERT · NLP · YouTube Data API
🏨 Royal Stay HMS Full-stack Hotel Management System covering bookings, check-ins, check-outs, and guest records. JavaScript · Flask · SQL

πŸ“Œ Explore all my work on the Repositories tab.


πŸ“Š GitHub Stats

GitHub Streak

Profile Summary

Repos per Language Most Commit Language

Activity Graph


πŸ“š Research & Writing

Beyond the Model: The Missing Layer in AI-Driven Systems A look at AI reliability as an infrastructure problem β€” arguing that the wrapper around the model often matters more than the model itself. πŸ“„ Published on ResearchGate Β· DOI: 10.5281/zenodo.20759186


🀝 Let's Connect

"Any sufficiently advanced technology is indistinguishable from magic β€” but engineers know where the magic breaks."

footer

Pinned Loading

  1. Intelligent-Youtube-Comment-Analyzer Intelligent-Youtube-Comment-Analyzer Public

    Uncover the true pulse of your audience. This web-based tool applies machine learning to user comments, delivering actionable insights that empower you to improve user engagement effectively.

    HTML 6