Available for engagements · 2026

Muhammad Ahmed

AI Operations Specialist · Engineering Student at SSUET Karachi. I design and operate multimodal AI systems — text, image, video, and music — with a relentless focus on truth, transparency, and operational reliability.

BasedKarachi, PK
StudioAI Operations
StandardHalal · Transparent
Enter the Workbench
01 — The Ethical Standard

Built on truth. Operated with integrity.

I am committed to Halal business practices. I do not engage in projects involving Riba (interest), gambling, alcohol, or deceptive AI marketing. My work is built on 100% truth and transparency.

What I commit to

  • Honest scoping — clear deliverables and clear limits.
  • Transparent AI — disclosure of generated content where it matters.
  • Reliable operations — measurable, auditable workflows.
  • Fair pricing — no hidden fees, no interest-based instruments.

What I decline

  • Riba (Interest)No interest-based financial products.
  • GamblingNo betting, wagering, or chance-based products.
  • AlcoholNo alcohol marketing, sales, or supply chain work.
  • Deceptive AI MarketingNo fake reviews, manipulated media, or dark-pattern funnels.
02 — Interactive Workbench

Live tools. No theatre.

Two demo agents you can interact with right now. Both are deliberate, constrained, and honest about what they don't know.

Module 01

PDF Intelligence

Drop a document. The agent extracts entities, summarises sections, and surfaces risk in plain language.

Drop a PDF here, or browse

Max 25 MB · Demo only

Local processing demo
v1.2
Module 02

Truth-Check Agent

Paste any claim. The agent rates plausibility, flags rhetorical patterns, and recommends what to verify.

0 chars
03 — About

An engineering student in Karachi, building the operations layer for multimodal AI.

StudyingBS CS — SSUET
CityKarachi
FocusAI Ops
StackMultimodal

I'm Muhammad Ahmed Chishti, an engineering undergraduate at Sir Syed University of Engineering & Technology in Karachi. My days are split between coursework, late-night builds in my home studio, and shipping operations for clients who need AI that actually behaves.

I work across the full multimodal stack — language models for research and automation, diffusion models for image and video, and generative audio for music and sound design. The unifying thread is operational reliability: taking flashy demos and turning them into systems that hold up under real load and real scrutiny.

Karachi taught me to be resourceful. SSUET taught me to be rigorous. Together they shape how I work — small, focused teams, honest scopes, and a refusal to ship things I wouldn't use myself.

01
Text
RAG, agents, structured extraction
02
Image
Generation, editing, vision pipelines
03
Video
Scene-aware editing & synthesis
04
Music
Composition, stems, sound design
04 — Skills

Three disciplines. One operating system.

01

Process Automation

Mapping operational bottlenecks and replacing them with reliable, observable AI-powered workflows.

  • Zapier · n8n · Make
  • Custom orchestration
  • Human-in-the-loop
02

Custom AI Agents

Purpose-built agents with tight tool boundaries, transparent prompts, and predictable cost profiles.

  • RAG pipelines
  • Tool-using agents
  • Eval-first delivery
03

Multimodal Workflows

End-to-end pipelines stitching language, vision, video, and audio models into one production system.

  • Text · Image · Video
  • Generative audio
  • Asset pipelines