I'm Shahzaib, a Full-Stack AI Engineer based in Pakistan.
I build production AI systems end-to-end: RAG pipelines, multi-agent workflows,
voice AI agents, and the backend and frontend that ships them to users.
As a founding engineer across hospitality, legal, governance, and SaaS,
I've led teams, owned roadmaps, and shipped from zero to production.
Beyond writing code, I do product research, write technical specs, and
translate them into sprint plans. I'm comfortable owning a product end-to-end
or plugging into an existing team as a technical lead.
Outside of work, I'm driven by my faith. It shapes how I approach building, leading, and solving problems.
What I'm building
Simplora.ai
: AI-powered meeting management platform that automates the complete meeting
lifecycle with pre-meeting briefs, real-time summaries, and automatic action
item extraction. Scaled to 300+ active users.
DreamDesk AI
: Enterprise AI for hospitality with multi-channel agents handling phone calls,
emails, web chat, WhatsApp, and Booking.com. Integrates with Mews and Guestline
PMS for real-time reservation sync.
Educational Background
I studied Computer Science at NUCES (FAST), graduating in 2024. During
my studies I co-authored research at the intersection of AI and software engineering.
Research Project Details:
LLM-assisted High-Quality Property Generation for Solidity Smart Contracts
under review
Under review at STVR · Collaborated with Dr. Hassan (DFKI Bremen) & Dr. Jalil (UAEU) · 2024
- Proposed an LLM-assisted pipeline to generate high-quality test invariants for property-based testing of Solidity smart contracts.
- Built an intelligent agent to automatically resolve compilation errors in LLM-generated properties.
- Evaluated prompt engineering techniques and fine-tuned hyperparameters to reduce hallucinations.
- Benchmarked via mutation testing, LLM-assisted approach achieved a 25.99% mutation score vs 31.75% manual, demonstrating meaningful reduction in human effort.
LLMSolidityProperty-based TestingMutation TestingLangChainAzure MLAI4SE
Stack & how I work
I work across the full stack and own deployment end-to-end:
- Backend: FastAPI, Node.js, gRPC, REST
- Frontend: Next.js, React, TypeScript
- AI / Agents: LangChain, LangGraph, LlamaIndex, RAG, Voice AI (LiveKit, VAPI), Pydantic AI, AWS Bedrock, MCP servers
- Infra & DevOps: Docker, GCP, AWS EC2/S3, Nginx, GitHub Actions, Digital Ocean
- Databases: PostgreSQL, MongoDB, Redis, Supabase, Pinecone, pgvector