The creative process began with a clear directive: produce a high-energy, visually immersive sizzle that embodied SHRED’s brand of cutting-edge AI innovation. During the initial development phase, I collaborated with the internal team to define the tone - cinematic, kinetic, and technically forward — while maintaining a balance between abstraction and clarity. I sketched out rough storyboards and mapped a visual rhythm that built from atmospheric intrigue to an explosive payoff, anchored by punchy motion design and a curated AI aesthetic. From there I utilized Google Whisk to generate high-quality visual references and character descriptions that would later form the foundation of the video. The planning phase also included research into mood references, pacing from reference trailers, and establishing a modular structure for future deliverables.

Once the narrative and structure were in place, I moved into asset creation, leaning heavily on AI tools to bring unique visual textures and transitions to life. Platforms like Runway, Veo3, ComfyUI, Krea AI, and Midjourney were used extensively, with tailored prompt engineering guiding the outputs toward the desired visual identity. These sequences ranged from surreal environmental loops to abstract glitch-driven imagery, which I then composited, masked, or layered with other motion elements to enhance depth and cohesion.

In parallel, I created custom motion design sequences in After Effects, employing kinetic type, particle systems, and screen replacements to build momentum between AI-driven segments. Each element—whether human-made or AI-generated—was treated with the same visual precision, blending seamlessly into the larger visual system through color grading, motion matching, and VFX polish.

For audio-driven scenes, I used ElevenLabs to prototype voiceover concepts for SHRED’s fictional AI coach, the central hero character, along with the trainer character in the final video. I had to carefully retime dialog to sync with Harvey's visual reactions. This included coaching lines like “Let’s work. Dumbbell Press first,” and user replies like “Yeah, let’s go.”

Cleanup/upscale: Topaz Video AI for denoise, de‑flicker, and resolution enhancement on strong but soft clips.

Fixes/inpaints: Runway for quick generative cleanup (wall artifacts, plate gaps); AE paint/masks for precise cosmetics.

Color pipeline: Temp treatment in Runway for curation; final look in Premiere/AE (contrast‑first, subtle filmic curve, brand‑true palette).

Voice prototype: ElevenLabs for AI‑coach VO timing passes (“Form looks off—drop 10% weight and slow tempo”).

After assembling a first batch of usable generated AI scenes, I moved into a rough cut and quickly flagged the clips that needed AI cleanup, upscaling, and denoising. For visually strong but slightly soft or noisy shots, I ran them through Topaz Video AI, which helped restore clarity, reduce flicker, and improve resolution.

Runway became a quick tool for previewing color treatments and testing generative inpainting, especially when filling in wall gaps or correcting model glitches. For more distracting artifacts—like distorted body contours or odd render details—I shifted into After Effects, using Mocha AE to motion track and digitally paint out the unwanted elements. This combination of automated and manual fixes allowed me to refine the footage while keeping the overall look cohesive and clean.

Once the final visual was approved, I exported the high-resolution master (ProRes) and then encoded optimized versions for social and web. Delivery included the full-length sizzle as well as modular assets for future promotional use, from vertical teasers to looped transitions and isolated typography pulls. Each version was carefully tested for compression artifacts and playback fidelity across devices.

In addition to the rendered files, I handed off a clean and organized project structure, with labeled comps, documented AI source prompts, and a folder of reusable motion elements that can be carried into future SHRED campaigns. This handoff empowers the team to scale the system I built, using both traditional and AI-driven assets in future content. The SHRED AI Sizzle 3 project stands as a showcase of how generative tools and expert post-production can coexist to deliver work that is both experimental and executable.