Skip to content
View SAMithila's full-sized avatar
🐢
Practising
🐢
Practising
1 follower · 1 following

Achievements

Achievement: YOLO

Achievements

Achievement: YOLO

Block or report SAMithila

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
SAMithila/README.md

Hi, I'm Samia 👋

AI Engineer focused on building production-grade AI systems including LLM agents, Retrieval-Augmented Generation (RAG) pipelines, and computer vision applications.

🎓 Master of AI graduate (Australia Awards Scholar) specializing in Natural Language Processing
🔬 Currently building AI systems combining LLMs, search, and vision models
📍 Interested in AI/ML Engineer roles in Singapore, UAE, and Remote

Featured Projects

LLM SQL Agent

Natural language → SQL database agent that allows users to query databases using plain English.

Highlights:

  • Built with LangGraph and guardrails
  • Schema-aware SQL generation
  • Deterministic execution pipeline
  • 96% benchmark accuracy

🔗 Repository

RAG Document Intelligence

Production-grade Retrieval-Augmented Generation system for querying document collections.

Features:

  • Hybrid retrieval (vector + keyword search)
  • Query expansion (HyDE)
  • Hallucination detection
  • 74% accuracy across 38 evaluation queries

🔗 Repository

LLM API Gateway

Unified backend for multiple AI providers powering 5 AI products.

Capabilities:

  • Single API for multiple LLM providers
  • Automatic failover (Groq → Gemini → OpenAI)
  • Real-time cost tracking
  • Session management (in-memory + Redis)

Supports: OpenAI | Gemini | Groq

🔗 Repository

Object Detection + Tracking Pipeline

Computer vision system for real-time object detection and tracking.

Tech:

  • Mask R-CNN for detection
  • SORT tracking algorithm with Kalman filtering
  • Self-supervised evaluation metrics
  • 78.4% tracking accuracy with 100% ID stability

🔗 Repository

AI Systems Expertise

Area Skills
LLM Applications RAG, LLM Agents, Tool Use, Function Calling
Search & Retrieval Hybrid Search (Vector + BM25), Query Expansion, ChromaDB
Prompt Engineering Few-shot, Chain-of-Thought, System Prompts
Computer Vision Object Detection, Tracking, Medical Imaging, Foundation Models (SAM)
Infrastructure API Orchestration, Multi-provider Failover, Session Management

Tech Stack

Category Technologies
Languages Python
AI/ML PyTorch, TensorFlow, Transformers, LLMs, NLP, Computer Vision
LLM Ecosystem OpenAI, Anthropic, Groq, Google Gemini, LangChain, LangGraph
Backend FastAPI, REST APIs, Redis
Tools Docker, Git, CI/CD (GitHub Actions), VS Code, Jupyter

Let's Connect

LinkedIn Email

⭐ Open to AI/ML Engineer opportunities

Pinned Loading

  1. rag-document-intelligence rag-document-intelligence Public

    Production-grade RAG system with hybrid search, query expansion, and hallucination detection. 74% accuracy on 38 queries across 5 documents.

    Python 1

  2. llm-api-gateway llm-api-gateway Public

    Unified AI API Gateway - One API powering multiple AI products with automatic failover, cost tracking, and session management. Supports Groq, Gemini, and OpenAI.

    Python 2

  3. object-detection-tracking-pipeline object-detection-tracking-pipeline Public

    Production-quality object detection & tracking pipeline with self-supervised evaluation metrics. Built with Mask R-CNN + SORT algorithm.

    Python 1

  4. llm-sql-agent llm-sql-agent Public

    A deterministic NL→DB agent — query any database in plain English with LangGraph, guardrails, and 96% benchmark accuracy

    Python