Mark Barnes / MarkVizion

AI Systems Experiments

Research tracks behind MarkVizion systems work.

Agent Environmental Awareness

Can a local AI agent become more useful when it sees desktop state before choosing actions? Compare text-only prompting against a loop that includes screen context, tool availability, and post-action evaluation. Reduced repeated explanation, better task continuity, and more reliable routing from intent to action. Next: Add stronger benchmark tasks with before/after screenshots, task time, and correction counts.

Animation Thinking for AI UX

Can animation principles make AI systems easier to trust and understand? Apply staging, timing, anticipation, and feedback to AI interface states instead of treating outputs as static responses. Users can better understand what the system is doing, what changed, and what to do next. Next: Create side-by-side demos of hidden-state AI output versus staged AI workflow output.

Case Study Authority Compounding

Does every shipped artifact become more searchable when paired with a demo, transcript, case study, and paper? Publish product pages, transcript summaries, structured case studies, and framework pages around the same project cluster. More unique query surfaces for Mark Barnes + OnyxKraken, Mark Vizion + local AI, and Creative Systems Engineering. Next: Track impressions and indexed pages in Google Search Console after deployment.