Mark Kashef
How I 10X My Productivity with ONE Claude Code Trick
Mark Kashef claims a Claude Code "assembly line" of three natural-language sub-agents makes you up to 10x more productive, producing a 25-slide researched presentation in about 10 minutes.
MK
Mark Kashef
Creator and host
14:09
The numbers, across the talk6 moments · 14:09
01 · AI sub-agent assembly line for non-coding knowledge work
Instead of manually copy-pasting between chat sessions, Kashef sets up a chain of Claude Code sub-agents that research, draft and edit autonomously. He claims this makes a user up to ten times more productive, with no coding required.
00:00:37
▲up to 10x more productive
Productivity
Productivityprojected
“So if you watch till the end, then you'll be able to create your own mini AI workforce that can execute tasks for you, and more importantly, make you up to 10 times more productive.”
02 · "State of AI Q4 2025" slide deck built by a three-agent pipeline
A research analyst, creative copywriter and senior editor agent were chained in Claude Code to produce a 25-slide presentation for a non-technical business audience. The agents executed and baton-passed their subtasks in roughly ten minutes, delivering final slides plus all intermediate research and draft assets.
00:01:12
●3 agents
Scale
Agents in pipeline
“So if we take a look right here, we have three agents.”
00:01:51
▲25 slides
Scale
Slides produced
“And you'll see the presentation basically says, create content for a 25 slide presentation titled the State of AI Quarter 4 2025 for a non-technical business audience.”
00:03:12
▼~10 minutes
Cycle time
Run time for full deck
“They're all invoked, and behind the scenes, they took around 10 minutes to go all execute their subtasks, then baton pass it to the next one.”
03 · Content calendar agents for a real estate AI workshop
Using his "assembly line" mega-prompt in Gemini, Kashef generated two new sub-agents (a real estate AI strategist and an instructional designer) plus a starter prompt, then created them in Claude Code. He estimates a run of this kind takes five to twenty minutes depending on the task.
00:12:02
●2 new agents
Scale
Agents generated
“So we'll take on copy and then you can see right here, agent changes. We now have two brand new agents.”
00:12:48
●5–20 minutes
Cycle time
Expected task run timeestimated
“This will probably take anywhere from five to 20 minutes, depending on the task, but the assets will look very similar to what you saw before.”