Skip to content
BackNov 7
Shake Khadija K.

Shake Khadija K.

AI Workflow & Automation Architect | AI Agents, Claude, Generative AI

I design AI workflows that run a business, not a demo: agents, automations and generative video pipelines, with a human check before anything ships. I'm the founder of Trivo, an AI-native studio. In 6 months we delivered 6,000+ Generative AI videos to clients in all 50 US states, Canada, Mexico and Australia, growing from 300+ a month to 1,000+ with next-day delivery. I designed the production system behind that and lead a team of amazing and creative professionals that runs it. Before AI, I spent 7+ years running marketing and content operations for 40+ brands. That taught me which work is worth automating: the tasks a team repeats every day. WHAT WE'VE BUILT THIS YEAR • AI video pipeline (prototype): an order ID goes in and a finished 30-second 1080p video comes out. 9 stages, AI vision, automated rendering, 25 automated QC checks and human approval before anything publishes. 2,100+ automated tests. • 3 custom MCP servers / Claude plugins that connect Claude to an admin API, Google Sheets and Frame io. They run in dev by default, production needs an explicit switch, and secrets never sit in the code. • Browser-agent automation that replaced a manual delivery workflow (Sheets → Frame io → admin portal). First full run: 5 of 5 orders, zero failures. Broken links get flagged instead of failing silently. • A prompt system that gets image-to-video models (Kling, Seedance, Higgsfield) to produce realistic camera motion from still photos, plus the QC playbook our editors follow. One rule outranks the rest: nothing about the subject gets invented. • AI virtual staging that keeps furniture consistent across every camera angle of a room. • AI presenters, voiceover and music for short-form video. WHAT I RUN • The production system: intake, tracking, status rules, SLA recovery and human review on every video • A distributed team of AI video editors, a project manager and a tech lead, working async-first • Model testing: we try new video and LLM models the week they launch and move only proven ones into production TOOLS Claude (Code, Cowork, skills, MCP) · ChatGPT · Gemini · Kling · Seedance · Higgsfield · ElevenLabs · Google Apps Script · Zapier · Frame io · ClickUp (certified Power User) · Notion · Premiere Pro · DaVinci Resolve HOW I WORK • AI for speed, humans for taste. Every system gets a review gate, a budget cap and a test run before it touches live work. • Numbers, not adjectives. If I say a workflow works, there's a run log to show it. • Async-first: written updates, one named point of contact, replies within a few hours. GOOD FITS • A task your team repeats every day across Sheets, Drive, a CRM or an admin portal • Connecting Claude or ChatGPT to your own tools and data • AI video or images at volume, with quality control a human signs off • Rolling out AI in a team: designing the workflow, training people and keeping it running • Marketing and content teams that want AI to carry the repetitive work Native Bengali, fluent English. Working async with US clients since 2020. Tell me the task your team repeats every day. I'll tell you how I'd automate it and what it would take.

🇧🇩Chattogram, Bangladesh$45/hr100% JSS4.90 (29)Since Apr 2020
GlobeUpwork

Earnings and projects

MRR
$0

Last 6 months ÷ 6

World#21,348 / 412k
🇧🇩Country#463 / 15.0k
Revenue, last 30 days
$0
Total earnings
$160k

Billed in all: $167k

World#21,348 / 412k
🇧🇩Country#463 / 15.0k
Per project
$3.9k
Projects, last 6 months
0

Across 2 contracts

0% fixed100% hourly
World#209,850 / 412k
🇧🇩Country#8,489 / 15.0k
Projects in all
41

Across 57 contracts

40% fixed60% hourly
World#74,019 / 412k
🇧🇩Country#5,462 / 15.0k

Earnings, last 30 days

$167kNothing earned in the last 30 daysBilled in all, before Upwork's fee · drag across the chart to see what was earned in any stretch
$164k$166k$168k$170kOct 8Oct 23Nov 7

Skills