Muhammad Adnan

Muhammad Adnan

About

Detail

AI/ML Engineer | Agentic AI & LLM Systems (LangChain, LangGraph, RAG) | AWS Data Engineering | FastAPI, Snowflake, Vector DBs
Lahore, Punjab, Pakistan
Contact Muhammad regarding: 
work
Full-time jobs

Timeline

work
Job
school
Education
folder
Project

Résumé

Jobs verified_user 0% verified
  • Axiom World
    AI Sales Agent for ERP Order Management
    Axiom World
    Jan 2026 - Current (10 months)
    LangGraph, LangChain, Mistral LLM, OpenAI GPT, FastAPI, Odoo V19, Python

    Automated the full sales email workflow for a manufacturing ERP: processing inbound customer enquiries, generating quotations,
    confirming orders, and drafting replies autonomously.
    Built a multi-stage, stateful agentic pipeline (Email Intake → Classification → Sub-Agent Routing → Odoo Writeback) with
    configurable LLM backends and rule-based security guardrails to prevent malicious emails from manipulating agent actions.
  • Axiom World
    Document Classification & Reasoning Agent
    Axiom World
    Sep 2025 - Dec 2025 (4 months)
    LangGraph, Amazon Bedrock, OCR (PDF/DOCX), PostgreSQL

    Automated document understanding by classifying files into database-driven subcategories with LLM-based reasoning.
    Built a multi-stage AI pipeline (OCR → Summarization → Classification) with iterative summarization and stateful graph
    orchestration.
  • Axiom World
    Council Question Answering System
    Axiom World
    Jun 2025 - Aug 2025 (3 months)
    FastAPI, LangChain, ChromaDB, PostgreSQL, OpenAI Embeddings

    Delivered accurate, context-aware answers from council documents using natural language queries and Retrieval-Augmented
    Generation (RAG).
    Built context-aware document retrieval with conversational memory maintained across multiple councils.
  • Axiom World
    AI/ML Engineer
    Axiom World
    May 2025 - Current (1 year 6 months)
    • Built an AI Sales Agent for ERP order management using LangGraph and configurable LLM backends (Mistral, OpenAI), automating the inbound sales email workflow end-to-end — classification, quotation generation, order confirmation, and reply drafting — integrated directly with Odoo V19
    • Designed and implemented multi-agent, agentic AI workflows to automate decision-making and improve data accessibility across business processes
    • Built a Course Recommender Agent using content-based filtering and semantic similarity search
    • Built a Document Classification & Reasoning Agent combining NLP and OCR for automated, database-driven document categorization
    • Developed a Council AI Chatbot using Retrieval-Augmented Generation (RA
  • Axiom World
    Junior Data Engineer
    Axiom World
    Feb 2025 - May 2025 (4 months)
    • Designed and scaled ETL/ELT pipelines feeding into Snowflake for downstream analytics
    • Delivered secure, scalable cloud data solutions using AWS S3, Lambda, Secrets Manager, IAM, API Gateway, Glue, and Custom Lambda Layers
    • Conducted Exploratory Data Analysis (EDA) on election datasets using Chi-Square Test, Two-Proportion Z-Test, and Cramér's V to identify statistically significant relationships
    • Collaborated on integrating data pipelines with downstream AI/ML systems
  • Axiom World
    LLamaIndex
    Axiom World
    Jan 2025 - Jul 2026 (1 year 7 months)
    Work with it's alternative framework LangGraph
  • Axiom World
    Data Enginner Intern
    Axiom World
    Nov 2024 - Jan 2025 (3 months)
    • Performed data extraction, cleaning, and transformation using Python and SQL
    • Designed and maintained ETL pipelines to extract, transform, and load data into Snowflake
    • Built automated data pipelines using AWS Glue for cross-service integration
    • Utilized AWS services — S3, Lambda, IAM, and CloudWatch — to support data workflows
  • Algo
    Data Science Intern
    Algo
    May 2024 - Aug 2024 (4 months)
    • Supported data-driven solutions for supply chain analytics and chatbot integration, organizing and integrating company data for interactive querying and reporting.
    • Performed data preprocessing, feature engineering, and model evaluation across multiple ML projects.
    • Contributed to NLP and computer vision workstreams, including SRT extraction and image classification.
  • Xpert Prime
    Business Development Intern
    Xpert Prime
    Feb 2024 - Apr 2024 (3 months)
  • Cplus Soft
    AI Intern
    Cplus Soft
    Aug 2023 - Sep 2023 (2 months)
    • Recommendations for crops based on factors like nitrogen, phosphorus, potassium levels, temperature, humidity, pH, and rainfall.
    • Generate and augment a dataset of 60,000 images, computing labels, and setting up a conditional GAN (CGAN) for image generation.
    • Object detection using YOLOv3, achieving accurate results in identifying objects within images.
    • Contributed to data-driven projects by implementing machine learning models and performing data preprocessing.
    • Generated actionable insights to support business decisions and improve operational efficiency.
    • Collaborated on tasks including data cleaning, feature engineering, and model evaluation.
Education verified_user 0% verified
  • International Islamic University Islamabad
    Bachelor's degree in Computer Science, Computer Science
    International Islamic University Islamabad
    Sep 2019 - Aug 2023 (4 years)
Projects (professional or personal) verified_user 0% verified
  • Axiom World
    AI Sales Agent for ERP Order Management
    Axiom World
    Jan 2026 - Current (10 months)
    LangGraph, LangChain, Mistral LLM, OpenAI GPT, FastAPI, Odoo V19, Python

    Automated the full sales email workflow for a manufacturing ERP: processing inbound customer enquiries, generating quotations, confirming orders, and drafting replies autonomously.
    Built a multi-stage, stateful agentic pipeline (Email Intake → Classification → Sub-Agent Routing → Odoo Writeback) with configurable LLM backends and rule-based security guardrails to prevent malicious emails from manipulating agent actions.
  • Axiom World
    Document Classification & Reasoning Agent
    Axiom World
    Sep 2025 - Dec 2025 (4 months)
    LangGraph, Amazon Bedrock, OCR (PDF/DOCX), PostgreSQL

    Automated document understanding by classifying files into database-driven subcategories with LLM-based reasoning.
    Built a multi-stage AI pipeline (OCR → Summarization → Classification) with iterative summarization and stateful graph
    orchestration.
  • Axiom World
    Course Recommendation Chatbot (LLM + RAG System)
    Axiom World
    May 2025 - Aug 2025 (4 months)
    Python (Pandas, SciPy, Matplotlib), Chi-Square Test, Z-Test, Cramér's V

    Analyzed close electoral races and identified shifts in party performance over a decade using statistical hypothesis testing.
    Revealed significant post-2016 trend shifts and high-volatility swing districts through margin-based close-race analysis.