Skip to content

Kayam PathanAI Engineer

Building intelligent systems with Generative AI, LLMs, RAG & AI Agents.

I'm an AI Engineer with an MSc in Artificial Intelligence from Queen Mary University of London. I build RAG pipelines, LLM-powered chatbots and autonomous agents in Python, and take them from data pipeline to deployment.

View ProjectsContact MeDownload Resume

Shipping GenAI solutions end to end.

I'm an AI Engineer with an MSc in Artificial Intelligence from Queen Mary University of London, and hands-on experience building Generative AI and Large Language Model systems.

My work covers RAG pipelines, AI-powered chatbots and autonomous AI agents, built with Python, SQL, LangChain, vector databases and prompt engineering, on a foundation of NLP, machine learning and deep learning.

I focus on shipping end-to-end GenAI solutions, from the data pipeline through to deployment.

LLMsRAGAI AgentsMCPNLPPythonSQLLangChainVector DBPrompt EngineeringHugging FaceMachine LearningDeep LearningChatbotsAWSDocker

Where I've applied it.

  1. Feb 2026 – Aug 2026

    Remote, India

    Generative AI Intern

    BharatSkillz

    • Independently engineered an Intelligent Customer Support Chatbot using LLMs and NLP, automating responses to routine queries.
    • Designed prompt engineering strategies and conversational flows to improve response quality and user experience.
    • Integrated chatbot logic with backend APIs in Python for real-time, context-aware query resolution, reducing support team workload.

    30%

    Reduction in manual handling time

    18%

    Improvement in intent recognition accuracy

    500+

    Daily interactions supported

  2. Mar 2022 – Aug 2022

    Nashik, India

    AI/ML Research Intern & Data Analyst

    Application Square Infotech

    • Developed and evaluated machine learning classification models.
    • Analyzed customer feedback with NLP techniques, surfacing insights that shaped key product improvements.
    • Automated data preprocessing workflows with Python and Pandas.

    Up to 88%

    Classification model accuracy

    1,000+

    Customer feedback entries analyzed with NLP

    50%

    Reduction in data preparation time

Systems I've built.

Three independent projects, each taken from data to a working system. Open one for the problem, architecture, stack and results.

RAGIndependent project

Financial Document Intelligence Platform

A Retrieval-Augmented Generation pipeline over real SEC filings (10-K and 10-Q) that answers natural-language questions with cited, source-level answers.

8

public companies covered

96.6%

exact match against SEC XBRL data

  • Primary/fallback LLM architecture
  • Live demo on free-tier infrastructure
  • RAG
  • SEC 10-K / 10-Q
  • Sentence transformers
  • FAISS
  • LLM APIs
  • XBRL validation
  • Gemini API
  • Ollama
  • Streamlit Cloud
SEC Filings1/8
MCPIndependent project

Autonomous IT Support Agent

An AI agent on the Model Context Protocol that autonomously triages, categorizes and routes IT support tickets, and logs every resolution for SLA tracking.

40%

reduction in triage time

25%

better semantic search accuracy

5+

SLA metrics in Power BI

  • MCP
  • AI agent
  • RAG
  • LangChain
  • Hugging Face
  • SQL
  • Power BI
  • Docker
  • AWS
Ticket1/7
n8nIndependent project

AI Job Application Copilot

An autonomous n8n workflow that matches a resume profile against live job postings using Gemini scoring, then generates cover letters and interview questions.

250+

listings scored per run

43%

qualified match rate

  • n8n
  • Gemini
  • Job board APIs
  • Structured JSON prompting
  • OAuth2
  • Google Sheets
  • Cover letters
  • Interview questions
Job APIs1/8

The toolkit, mapped.

Pick a group to see its skills orbit. There are no percentages here on purpose: every skill listed is one I've used in the work above.

GenAI & LLMs

How I build AI systems

From raw data to a deployed system, these are the stages I work through and the tools from my projects that sit in each one. Hover or tap a stage.

Step 1 of 8

Data

Real sources rather than toy data: SEC 10-K and 10-Q filings, internal IT runbooks, live job board APIs and 1,000+ customer feedback entries.

  • SEC filings
  • Runbooks
  • Job board APIs
  • Customer feedback

Where I studied.

  1. 2024 – 2025

    London, United Kingdom

    MSc Artificial Intelligence

    Queen Mary University of London (Russell Group University)

    GPA 3.7/4.0

  2. 2020 – 2024

    Nashik, India

    B.E. in Computer Science & Engineering

    Savitribai Phule Pune University

    Distinction

Have an AI problem worth solving?

Let’s build something intelligent.

Send an email