AI Engineer with 1+ years in LLMs, Multi-Agent Systems, RAG pipelines, and agentic frameworks with 1
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI Engineer specializing in LLMs, Multi-Agent Systems, RAG pipelines, and agentic frameworks. Experienced in building production-ready intelligent systems using LangChain, LangGraph, Google ADK, and Model Context Protocol (MCP). Skilled in context engineering, AI evaluation, computer vision, deep learning, and full-stack development. Proven track record of designing and deploying real-world AI solutions across retail, logistics, and generative UI domains.
University of Central Punjab (UCP)
Bachelor of Science · Artificial Intelligence
August 1, 2022 – June 30, 2026
Punjab College Hasipur
FSC Pre-Engineering
June 1, 2020 – May 31, 2022
Turing
Business Analyst & AI Contractor
March 1, 2026 – May 1, 2026
India
Digital Elites. Dev
Python Developer Intern
September 1, 2024 – December 1, 2024
Lahore, Punjab, Pakistan
Axiom AI - Neuro-Symbolic Financial Invoice Auditor
June 1, 2026 – Present
• Built a production-grade 5-agent neuro-symbolic RAG pipeline (Google ADK) that audits vendor invoices end-to-end: neural extraction of messy PDFs via a local LLM, exact symbolic verification (SymPy/Decimal) of every arithmetic relationship, and policy compliance through a deterministic rule engine • Engineered RAG over per-tenant policy documents (FAISS, bge-m3 embeddings) with layered hallucination controls: grounding gates, verbatim-citation enforcement, and severity capping so the LLM extracts and explains while symbolic engines decide correctness • Developed an offline, deterministic evaluation harness (9/10 pass) measuring overcharge-detection accuracy, grounding confidence, and hallucination-guard rejection; deployed as multi-tenant FastAPI + Next.js services
Agent-Based Inventory Management System
June 1, 2026 – Present
• Building an AI-driven inventory system using multi-agent architecture targeting Pakistan's retail and logistics sector • Autonomous agents handle demand forecasting, restocking alerts, supplier coordination, and anomaly detection across real-time data streams • Backend powered by FastAPI and PostgreSQL; Next.js frontend with real-time analytics dashboard
AI Shopping Advisor - Intelligent Multi-Agent Shopping System
June 1, 2026 – Present
• Built a 4-agent LangGraph pipeline handling product search, comparison, recommendation, and query resolution, tested across 50+ products • Achieved 80-85% recommendation accuracy with sub-30-second end-to-end response time across multi-step agent handoffs • Integrated FastAPI backend with SQLite and DummyJSON product API; Next.js frontend with conversational UI and memory persistence
Sales Navigator App - Fashion Discount Aggregator
June 1, 2026 – Present
• Built a web scraper aggregating 7,000+ live product listings across 10+ Pakistani fashion brands into a unified discovery platform • Developed real-time product feed with filtering, brand comparison, and UTM-tracked conversion links for performance analytics • Stack: Python (Scrapy/Requests), FastAPI, Next.js, PostgreSQL
AI-Powered Automotive Anomaly Detection
June 1, 2026 – Present
• Annotated 100+ images and built a structured computer vision dataset for automotive defect detection • Trained and evaluated YOLOv8 model achieving 68% baseline accuracy on a constrained dataset, with a documented improvement roadmap • Established evaluation criteria and quality guidelines for annotation and model assessment workflows
One Mix AI - AI-Powered UI Generation
June 1, 2026 – Present
• Built an AI system that generates fully structured UI for websites and mobile apps from a text prompt or hand-drawn sketch • Leverages multimodal LLMs for sketch-to-UI interpretation and context-aware component generation across web and mobile targets • Eliminates the gap between raw product ideas and production-ready interfaces without requiring design expertise
Atlas - Multi-Agent AI System
June 1, 2026 – Present
• Designing a production-grade multi-agent system using Google ADK and LangGraph with orchestrated agent communication, persistent memory, and tool use across multi-step workflows • Implementing MCP-based tool integrations and context engineering strategies for robust, stateful agent execution • Building an evaluation harness to benchmark agent performance across task completion rate, reasoning quality, and failure recovery
View ProjectLang Chain for LLM Application Development
DeepLearning. AI
June 1, 2026 – Present
Generative AI with Large Language Models
DeepLearning.AI / Coursera
June 1, 2026 – Present
Deep Learning Specialization
DeepLearning.AI / Coursera
June 1, 2026 – Present
Cultural Fit Analysis
The candidate's project portfolio demonstrates a strong interest in applying AI to diverse real-world problems, including financial auditing, inventory management, e-commerce, and UI generation. This breadth of application, combined with experience in both professional and academic settings, suggests adaptability and a proactive learning mindset. The focus on building production-grade systems aligns well with a results-oriented culture. However, the experience is relatively short, which might require mentorship in a senior role.
Soft Skills & Operational Fit
The candidate's professional experience as a Business Analyst & AI Contractor and Python Developer Intern indicates strong analytical skills, attention to detail in evaluating AI model outputs, and collaboration within cross-functional teams. The resume also highlights strong written and verbal English communication and comfort with remote work, which are crucial for operational fit.