Medium

Umair Ali Khan, Ph.D.

1.98K followers

Pinned

In

AI Advances

by

Umair Ali Khan, Ph.D.

·

Apr 27

How to Accurately Extract Everything from Documents Using AI

How asking the AI models to output in a specific format changes everything

How to Accurately Extract Everything from Documents Using AI

Pinned

In

Data Science Collective

by

Umair Ali Khan, Ph.D.

·

Feb 5

I Created a Claude Skill for Processing Purchase Orders

A no-code solution for multi-document processing

I Created a Claude Skill for Processing Purchase Orders

Pinned

In

Data Science Collective

by

Umair Ali Khan, Ph.D.

·

Jan 28

How to Extract Everything from Documents Using AI

I created a vision-enhanced document parser and RAG chunker

How to Extract Everything from Documents Using AI

Pinned

In

Data Science Collective

by

Umair Ali Khan, Ph.D.

·

Jan 2

I Built a Claude Skill for Knowledge Extraction & Report Writing

Claude Skill + MCP: extract knowledge from documents & media, and generate a report on your template

I Built a Claude Skill for Knowledge Extraction & Report Writing

Pinned

In

Data Science Collective

by

Umair Ali Khan, Ph.D.

·

Dec 4, 2025

How I Created a Claude’s “Skill” that Creates Full-Stack AI Applications

A step-by-step walkthrough of creating a custom Claude Skill to build full-stack applications

How I Created a Claude’s “Skill” that Creates Full-Stack AI Applications

In

AI Advances

by

Umair Ali Khan, Ph.D.

·

Oct 21

Layout-Aware Document Parsing for Reliable Structured Extraction with AI

AI can extract everything correctly and still lose the meaning. Here’s what to check before trusting its output.

Layout-Aware Document Parsing for Reliable Structured Extraction with AI

In

AI Advances

by

Umair Ali Khan, Ph.D.

·

Oct 20

I Built a Claude Skill for Accurate Knowledge Extraction & Report Writing

A normalize-curate-synthesize workflow that turns mixed-source material into long-form reports with traceable evidence.

I Built a Claude Skill for Accurate Knowledge Extraction & Report Writing

In

CodeToDeploy

by

Umair Ali Khan, Ph.D.

·

Oct 18

How Much Research Can We Safely Hand Over to AI Agents?

Where AI can help with research, grant writing, and peer review, and where human judgement remains essential.

How Much Research Can We Safely Hand Over to AI Agents?

In

AI Advances Briefs

by

Umair Ali Khan, Ph.D.

·

Oct 16

Your Smart Devices Could Become the Next Victims of Rogue AI Agents

Early experiments suggest small AI models may soon be able to copy themselves across cheap home devices.

Your Smart Devices Could Become the Next Victims of Rogue AI Agents

In

Data Science Collective

by

Umair Ali Khan, Ph.D.

·

Oct 9

Why Use an Unknown Model? How SenseNova U1.5 Quietly Challenges the Giants in Native 4K

I tested an open-source model on text-heavy design, complex scenes, and precise editing. The results were astonishing.

Why Use an Unknown Model? How SenseNova U1.5 Quietly Challenges the Giants in Native 4K

Umair Ali Khan, Ph.D.

Umair Ali Khan, Ph.D.

1.98K followers

Friend of Medium

Senior AI Researcher & AI Consultant | (AI/ML, LLMs, AI agents, structured knowledge extraction) Web:https://umairalipathan1980.github.io/resume/#

Following

Help

Status

About

Careers

Press

Blog

Store

Privacy

Rules

Terms

Text to speech