B.E. Computer System Engineering
DUET — strong foundations in systems & software.
Project Inquiries
For engagements, collaborations, and consulting.
Hello, I'm Zubair Khatti
AI Engineer focused on agentic LLMs, RAG, and knowledge graphs—delivering prototypes and production features that actually move the needle.
Design
Code
Ideas
AI Engineer
I build useful AI agentic LLMs, RAG systems, and real-time vision
2+ years in Computer Vision, now building GenAI: multi‑agent workflows (LangChain/LangGraph), retrieval with ChromaDB, and knowledge‑graph powered reasoning (Neo4j). Python + FastAPI, delivered with observability and iteration speed.
Recently shipped prototypes and production features across chatbots, retrieval pipelines, and AI surveillance—using pragmatic tools, strong data modeling, and clear success criteria.
LLM Agents
LangChain/ LangGraph, tool routing, eval loops.
RAG Systems
ChromaDB, embeddings, retrieval pipelines.
Knowledge Graphs
Neo4j, Cypher, Graph‑RAG patterns.
Computer Vision
YOLO, OpenCV, tracking, real‑time pipelines.
AI Engineer — Computer Vision
Face recognition, crowd monitoring, streaming, SQL‑backed logging.
Transition to GenAI
Agentic LLMs, RAG/ChromaDB, LangSmith tracing, Streamlit/Gradio.
AI Engineering & Consulting
Rapid prototypes → production features for real users.
Projects
A curated set of AI engineering work. LLM chatbots/agents, retrieval (RAG) pipelines, and computer‑vision demos. Click a card to open the repo or watch a short demo (private projects show demo only).
- All
- Chatbots
- Computer Vision
- Data Engineering
- Website
Event Booking Website
Modern website for event bookings
Skills
Core strengths across LLMs/agents, retrieval systems, knowledge graphs, and computer vision—grounded in Python and solid backend practices.
LLMs & Agents
LangChain, LangGraph, tool/function calling; model serving with Gemini 2.5 Flash, Qwen3‑30B (Fireworks), LLaMA 3.2 (Ollama).
Retrieval & Embeddings
RAG pipelines with ChromaDB; spaCy embeddings, mxbai‑embed‑large; evaluation and retrieval tuning.
Knowledge Graphs
Neo4j, Cypher, entity extraction; Graph‑RAG patterns for reasoning over connected data.
CV/ML
OpenCV, MTCNN, face_recognition; scikit‑learn (DBSCAN/ HDBSCAN/ MeanShift/ KMeans) real‑time analytics.
Also: Python, SQL, FastAPI, SQLAlchemy, REST APIs, Streamlit/Gradio, LangSmith tracing & evaluation.
Resume
View my resume or download the PDF.
Let’s build with AI
I help teams scope, prototype, and productionize AI features—LLM integrations, RAG systems, agents, and MLOps. Share a brief on your use case and I’ll get back within 24–48 hours.
Contact Info
For any inquiries, please contact me via email.
Preferred Contact
Email is best for initial scoping.
Happy to schedule a call after your message.
Location & Availability
Remote — Worldwide
Replies within 24–48 hours (business days)