Mark Christiansen

🎉 Jellyfish AI Impact now supports Windsurf, a leader in AI-powered coding solutions, offering tools like the Windsurf Editor that blend collaborative, continuous AI assistance with developers' workflows. Together, Windsurf and Jellyfish are changing how AI and humans work together.  Through our robust integration with Windsurf, Jellyfish AI Impact allows joint customers to:  🤖 Measure Windsurf’s impact on overall developer productivity: Analyze how Windsurf solutions like Windsurf Editor impact different engineering workflows and use cases – whether it's building new features, reducing technical debt, or reviewing for code quality and security improvements. 📈 Scale Windsurf adoption across the organization: Identify teams with high engagement to further accelerate development, automate testing, and streamline deployment, then replicate these best practices and implementation strategies company-wide. Learn more about what this integration means for your engineering organization here: https://lnkd.in/ghNguE47

54

Incidents like this are a strong reminder of how essential real, human QC is — even on rush jobs. A proper review on both sides would have caught this instantly. But in an age of automation and AI-driven workflows, mistakes like these can slip right through and derail all the hard work that should be the focus. Instead of celebrating the restoration or the quality of the show, everyone ends up talking about what should never have been visible in the first place. https://lnkd.in/gkX9SaPf

37

7 Comments

For my connections focused on #motioncapture and/or #metahumans, I thought I’d share a “sneak peak” at what appears to be some major enhancements to #unrealengine’s performance capture and digital human pipelines in UE 5.6.  Looks like the MetaHuman Creator system is not only moving into an in-engine tool, but also adding highly customizable body sizing controls and other major advances. The performance capture pipeline also appears to be retooled to be designed around the goal of centralizing live capture workflow, including custom performer->character mapping, session/slate pre-planning and session management. All of this is based on what I’m inferring from the public source code version of #urealengine5.6. The primary source of the best information on all of this will be the talks and developers at #unrealfest in Orlando next month. If you’re going, keep an eye out for me. I’ll be in the audience of all the MetaHuman and performance capture sessions can get into other than when I’m presenting my own lab session on the Motion Design toolset. Hope to see you in Orlando! https://lnkd.in/eaND2wJr

52

5 Comments

Please, consider not using AI voices, or internal resources for your concepts, pitches, animatics, etc... Let's find a way to use the same human talent (hopefully) you'll consider for the final!! Even the "union rate" for demo reads is pretty low, AND it gives you the chance to start the connection between talent, copy and client. If your client deserves YOUR creative talents... imagine the value added by a trained human artists! Cast them early! Have them standing by. Another tool in your toolbox, ready to make the project shine. In 2025 I worked on several projects that I was able to negotiate attractive demo-fees, so the producer could afford multiple adjustments or even full re-writes as we worked toward the final spot. When that final session date came, it was like old friends walking the path THEY CREATED! Smooth, tested and pre-approved! Adding VO doesn't have to be last minute every time. It can be early, and part of the creative process. 😉

510

148 Comments

The conversation around AI in filmmaking is loud. But the real work is quieter. It’s not about replacing artists. It’s about building better tools. Wētā FX’s workflow on Stranger Things 5 is a perfect case study. They integrated AI and AR not as a magic button, but as a direct extension of their artists' hands. This is the future of the studio pipeline—smarter, more iterative. But technology is only half the story. The artists still rely on physical references, like using BBQ steak to get the texture of monster gore just right. The core principles of the craft haven't changed. The most effective tools don't automate creativity; they sharpen it. Check the full article here: https://lnkd.in/gHQJ7dsc

15

4 Comments

I am determined to work with renewed vigour and purpose to “reimagine VFX & Animation in the AI-Native Era”! In the world of VFX, animation, and immersive storytelling, the direction of travel is clear: create faster, iterate smarter, and scale globally—without sacrificing quality. While individual AI tools now touch many parts of the pipeline: assets, rendering, etc.; the true breakthrough comes from integrating them into a cohesive, intelligent, efficient production stack. Collaborators, projects and those with AI agents are welcome to connect to us! What a group of us believe: 🎬 It’s time an AI-native CG content stack built for film, TV, games, XR, and location-based experiences will become a new normal ⚙️ Designs to streamline workflows, reduce cost & time, and potentially enable licensable modules—from smart asset libraries to AI-driven render layers will arrive sooner than we think 💡 We shall remain focused on creative empowerment, not replacement ✅ We’re merging production expertise with AI engineering to build the studio infrastructure of the future. Some things I mentioned in my talks in 2 countries over the past month: 💬 “The future of content isn’t just faster or cheaper—it’s smarter. AI unlocks a creative-industrial revolution where intelligence is embedded in every frame, every asset, every decision.” 💬 “Dozens of AI tools solve isolated problems. The real innovation lies in orchestrating them into a seamless, intelligent pipeline that redefines how premium content is made.” 💬 “We are best off not just building IP—we need to package intelligence. Our experiments and strategies are to create AI-native pipeline creates licensable tools, assets, and workflows that others can plug into their future.” 💬 “We are determined to collapse the time-cost equation in content creation. Hence, what once took months and millions should progressively be iterated in days, without compromising on quality or vision.” 💬 “AI isn’t here to replace human creativity—it’s here to supercharge it. The new global delivery model through AI is an enterprise stack we are determined to bring to the world to empower artists and producers to imagine more, iterate faster, and deliver better.” 🎯 Whether you're a VFX supervisor, animation lead, studio head, or investor—talk to us if you can help reshape a new frontier. #Iamonit #TheAteam #MovingToTheLaunchPad #VFX #Animation #CGPipeline #AIforArtists #GenerativeAI #MediaInnovation #VirtualProduction #RealtimeGraphics #CreativeTech #FilmTech #EntertainmentInvesting #AIStudio #ContentTech #ScalableProduction #FutureOfEntertainment #MediaandEntertainment #AIIntegration #LicensableIP #DigitalProduction #birenghosespeaks #inspiredbyAIleaders #CII4India #ABAI #GovernmentOfKarnataka #KTech

127

8 Comments

My quote in the new issue of VFX Voice is how I feel about education in general, not just around AI. “Educational institutions have a unique opportunity to learn from industry standards and histories while pushing the boundaries through emerging technologies,” notes Jimmy Calhoun, Chair of BFA 3D Animation and Visual Effects at The School of Visual Arts in New York City. “Our students understand this responsibility. They’re not only exploring the potential of AI but also reflecting on its impact on their rights as artists, the future of their mentors’ jobs and the environment.” https://lnkd.in/e6YemmX9

84

1 Comment

Character consistency in VEO3 I2V? The trick? The same start-frame for all clips and: "Instantly jump/cut on frame 1 to [describe the new scene loosely]. [Add a more detailed description of the new scene]. The song is "I'll Understand" by Roza, from http://Artlist.io. If your start frame has a chroma key background, VEO is more likely to quickly do the jump to the new scene you describe, while retaining the key elements from the start frame if you describe them loosely. It's not 100% but I still find it very useful when I'm creating clips. This trick works for other things than characters too..! :-) #Veo #genai

407

52 Comments

When a client asks for character animation, most beginners start keyframing every limb manually. But pros? They rig. In this tutorial, I broke down the process of using Duik Angela—a free tool inside After Effects—to build a custom skeleton and automate complex character movement using Inverse Kinematics (IK). What I covered: • What is Rigging? Creating a skeleton system that lets one body part influence the others—for example, when you move the arm, the hand follows naturally. • Planning the Character: I designed a custom character (“Karubhuddhi Lal Bhaiya in an MJ jacket”), broken into layers—arms, legs, torso, hands, etc.—to prep for rigging. • Setting up Duik Angela: Built a full bone structure using custom joints, linked everything together, and created a fully controllable puppet. • Animating with Ease: Used controllers to animate the dance sequence, with smooth movement and minimal manual keyframing. 

52

2 Comments

In 2026 half of my work is 3D and half is Gen AI (mostly hybrid). Unreal Engine really has the foundation to be an incredible way to control Generative AI. I've been working A LOT with the different Gen AI platforms that most have their own graph/node workflow but are pretty undercooked compared to Unreal Engine Blueprints, especially the new Control Rig Blueprints which are a generational leap forward from the traditional Actor Blueprints system. To continue to modernize Unreal Engine for the Generative AI era I've put together this simple previs graph. UE is great as coding and working with string aka "TEXT" (I know text this is it's own variable type too) and would make automating annoying text concats programmatically a breeze. Add a simple API call nodes to Google Nano Banana or Epic's own image model. Reference in CG cameras from Sequencer or at run time and have it output textures for clothing, faces, etc. at run time or for offline sequencer use.

354

18 Comments