Architecting Agentic
AI Systems.
Hi, I'm Sana Nasir. I specialize in bridging the gap between raw data and business intelligence by architecting autonomous agents that transform complex operations into streamlined workflows.
Technical Expertise
🔬 Gen AI & LLMs
LangGraph, Llama 3.3, Gemini 1.5/2.5 Pro, Groq API, Mistral, Hugging Face, LangChain, RAG Systems, Agentic Web-Intelligence.
👁️ Computer Vision
Tesseract OCR, OpenCV (Image Preprocessing), Document Intelligence, Vision-Language Models (VLMs).
💻 Engineering
Next.js 14, Python, FastAPI, TypeScript, Streamlit, Docker, PostgreSQL, SMTP Automation, Git/GitHub.
📊 Data & APIs
Pandas, NumPy, Scikit-learn, FAISS (Vector DB), Matplotlib, Seaborn, DuckDuckGo Search API.
🎯 Current Focus
Developing Dual-Inference benchmarking systems (Gemini vs Llama), multi-agent SEO orchestration, and high-velocity SaaS backends with FastAPI.
Featured AI Engineering Work
SEO_FORGE Pro
Global Content Intelligence engine pitting Gemini 3.1 vs Llama 3.3. Features a "Llama-as-a-Judge" auditor for SEO benchmarking and real-time TPS monitoring.
Receipt Physician Pro
AI financial auditor merging OpenCV/Tesseract with Mistral-7B for deep-dive audits, anomaly detection, and strategic savings plans.
View RepositorySalesScout AI Pro
Agentic B2B platform using Llama 3.3 and DuckDuckGo API to track real-time market signals and autonomous sales leads.
View RepositoryTalent Guard
AI Resume Ranking with batch-processing and automated rate-limiting. Generates AI "Executive Memos" for hiring insights.
View RepositoryLet's build the future.
Currently open to AI Engineering, Generative AI Specialist, and MLOps roles.