Uploaded August 2026 | Updated September 2026, 2 weeks ago
Every tech leader is talking about agentic development and the promise of dramatically speeding up software delivery. But as organizations begin scaling AI-generated code, many are running into the same challenge: their test data processes can't keep up.
In this Q&A, Perforce Delphix experts Matt Yeh (Director of Product Marketing) and Brian Muskoff (VP of Product) discuss what they're seeing in the market. Brian talks through the 4 main ways he sees agentic development initiatives failing, why test data has become a critical bottleneck, and what App Dev leaders need to rethink if they want to successfully scale AI-driven software delivery.
1:00 What are companies getting right about agentic development?
2:48 Why are the models NOT the problem?
4:09 Where does test data start slowing things down?
5:28 What does this bottleneck look like in the real world / where is agentic development getting stuck?
6:44 What does a better data foundation for agentic development look like?
9:06 Questions every App Dev leader should ask
#AgenticDevelopment #AI #DevOps #ApplicationDevelopment
---------------------------------------
See the key findings from our survey of 500+ enterprise leaders in our 2026 Test Data Management Report for AI-Ready Enterprises: https://ter.li/vw2t467y
Explore Perforce Delphix test data management solutions for agentic software delivery: https://ter.li/zaxq1r
Learn how Delphix Synthetic Data unblocks development for devs, testers, and agents: https://ter.li/e3yj3feo
Every tech leader is talking about agentic development and the promise of dramatically speeding up software delivery. But as organizations begin scaling AI-generated code, many are running into the same challenge: their test data processes can't keep up.
In this Q&A, Perforce Delphix experts Matt Yeh (Director of Product Marketing) and Brian Muskoff (VP of Product) discuss what they're seeing in the market. Brian talks through the 4 main ways he sees agentic development initiatives failing, why test data has become a critical bottleneck, and what App Dev leaders need to rethink if they want to successfully scale AI-driven software delivery.
1:00 What are companies getting right about agentic development?
2:48 Why are the models NOT the problem?
4:09 Where does test data start slowing things down?
5:28 What does this bottleneck look like in the real world / where is agentic development getting stuck?
6:44 What does a better data foundation for agentic development look like?
9:06 Questions every App Dev leader should ask
#AgenticDevelopment #AI #DevOps #ApplicationDevelopment
---------------------------------------
See the key findings from our survey of 500+ enterprise leaders in our 2026 Test Data Management Report for AI-Ready Enterprises: https://ter.li/vw2t467y
Explore Perforce Delphix test data management solutions for agentic software delivery: https://ter.li/zaxq1r
Learn how Delphix Synthetic Data unblocks development for devs, testers, and agents: https://ter.li/e3yj3feo






![Trading Render Farms for Real-Time at M2 Animation
At M2 Animation, scaling cinematic production meant rethinking everything—from rendering pipelines to how directors interact with scenes.
Benjamin Foo, CG Supervisor at M2, walks through the studio’s transition from a traditional Maya/Redshift workflow to a real-time, Unreal Engine-based pipeline. What started as a necessity—limited compute power and rising production demands—quickly became a creative advantage.
Instead of waiting on renders, teams now iterate instantly. Directors can step inside scenes, shape environments in real time, and collaborate more closely with artists. But that shift also brought new challenges—from managing massive asset libraries to balancing visual fidelity with hardware limits.
Ben shares how M2 builds custom tools, embraces experimentation, and uses Perforce P4 to manage version control across complex productions—all while keeping artists empowered to explore and create.
In this episode:
How M2 transitioned from a traditional render pipeline to Unreal Engine
Why real-time workflows unlocked faster iteration and creative freedom
The technical and cultural challenges of scaling cinematic production
How teams balance visual fidelity, performance, and hardware limits
The role of custom tooling and experimentation in pipeline evolution
How Perforce supports collaboration and version control across teams
Why empowering directors inside Unreal Engine changes the creative process
Timestamps:
[00:00:00] Cold Open
[00:02:25] Meet M2 Animation
[00:03:02] Ben’s Career Path
[00:04:15] CG Supervisor Role
[00:05:17] Before Unreal Pipeline
[00:06:00] Why Switch to Unreal
[00:07:21] First Unreal Trailer
[00:11:08] UDIMs and Virtual Textures
[00:14:05] VRAM Limits and Team Learning
[00:17:30] Directors in Real Time
[00:19:24] Asset Library and Versioning
[00:22:39] Shot Based vs Sequence Based
[00:25:34] USD vs FBX Today
[00:26:55] FBX vs USD Plans
[00:27:25] When Fixes Leave Unreal
[00:29:19] Material Notes Workflow
[00:32:32] Instances Per Shot Tweaks
[00:34:03] Perforce Locking Discipline
[00:35:43] Small Team Version Chaos
[00:39:48] Local Rendering Hardware Strategy
[00:42:23] Cinematic Look Pipeline
[00:45:39] Effects Mostly In Engine
[00:46:50] Lessons And Closing Advice
Links mentioned:
https://www.unrealengine.com/spotlights/shock-and-ork-how-m2-brought-a-warhammer-kill-team-battle-to-life?lang=en-US
https://youtu.be/MjjkWH0eT3U?si=eJlkriPPjwkV2_MP&t=298
https://youtu.be/p9LbaCcFpgE?si=KgjlDW1y0tCoZ9Ly
https://youtu.be/w-siBR9Kt2M?si=aeASyXGN6JQit2V4 Trading Render Farms for Real-Time at M2 Animation](https://i.ytimg.com/vi/qkzu4esUEH4/mqdefault.jpg)



