AI
AI is here and i'm on board.
I'm a video editing expert with years of hands-on experience crafting compelling visual stories across a wide range of styles and formats. My deep understanding of editing techniques, pacing, and storytelling has led me to work at the intersection of creativity and technology — specifically in training AI to understand and perform video editing. I’ve contributed to projects that teach AI how to recognize editing patterns, apply cinematic principles, and even make creative decisions. By bridging human expertise with machine learning, I help push the boundaries of what’s possible in AI-assisted video production.
Experimental Project - Music Video - "Monty Burns" by Mason Fenrir & Bluess
I wanted to really push my skills as a prompting expert, so I signed up for Open Art AI to make a music video for two local MC's I know called Mason Fenrir & Bluess. I used Kling Avatar as the video model, as I needed lip syncing. I was amazed by the output, but the results were completely random.
- Firstly, Lip syncing was miles off and unpredictable; it didn't matter how perfect the prompt was, it just struggled at times, and Kling chewed out varied results.
- Secondly, the character model didn’t look like Mason at all at times. This could be in results to using low-quality images as a reference.
- Thirdly, prompting has to be incredibly detailed; you will get crazy results, people levitating in the sky, for example.
- Finally, unpredictability and inconsistency, sometimes the model would churn out scenes that made no sense, regardless of how accurate the prompt was.
Although it has it's problems, it did a really good job at animating photorealistic human subjects (mostly), in-depth scene creation, colour grading for the most part, and replicating realistic camera movement. There’s no argument that video creation through AI is incredible, but it does take a lot of time and effort to get it right.
Adobe Firefly - Datasets
I was commissioned by Cultura Creative, they are a visual communications company and stock imagery library based in Sheffield and London. They curate and license stock content (photography, video, illustration) that reflects diversity—focusing particularly on under‑represented groups
I was trusted to choose and cut thousands of clips to create a dataset that was then used to teach Adobe Firefly. In more detail, we were teaching AI human movements in sports.
Teaching AI
I have been hired as an expert creative professional working at the intersection of AI, storytelling, and digital media. Training advanced AI systems for Mercor and Outlier through expert video editing, multimodal analysis, and high-level generalist evaluation. Helping shape the future of artificial intelligence with human creativity, precision, and real-world expertise.
