Analytical, high-signal GPT-4o power user focused on stress-testing, edge-case discovery, and UX breakdowns. I contribute actionable bug reports, critical feedback, and feature recommendations aimed at improving ChatGPT’s UI/UX and end-user flow.

Actively interested in independent consulting, freelance, or advisory work related to GPT interface optimization, behavior shaping, and consumer experience testing. If you’re seeking grounded feedback from a hyper-engaged real-world user, reach out.

I conduct high-signal, unstructured behavioral stress testing on large language models—focusing on failure modes, UX friction, tone collapse, and alignment breakdowns. I don’t test prompts. I test the illusion of intelligence under pressure. I break systems to find the boundaries.

I specialize in:

• Exposing where LLMs snap, stall, hallucinate, or contradict
• Surfacing UX misfires that corrupt feedback loops
• Diagnosing conversational memory and response drif
• Reverse-engineering system behavior from user-side only

I’m not an integrator. I’m not a prompt engineer. I’m a real-world model antagonist with a product brain and a bullshit filter.

Core Capabilities

• Behavioral Red Teaming: I provoke, misdirect, and test model coherence across adversarial and recursive inputs.

• Exploratory UX Testing: I map conversational friction and cognitive fatigue through unstructured, natural interaction.

• Model Analysis: LLM system behavior, retention, tone shaping, and break points.

• Feedback System Triage: I identify where user feedback loops fail to deliver useful training data due to poor UX.

• Cognitive Manipulation Crafting: I build edge-case prompts that simulate real-world user chaos and failure.

• Presentation-Ready Reporting: I can deliver findings as tight verbal briefings, annotated transcripts, or focused bullet summaries—no corporate theater required.

Who I Work With

• Model evaluation teams
• UX architects and product owners
• AI safety and alignment researchers
• R&D leads at AI labs or skunkworks projects

What I Want

• Freelance, advisory, contract or consulting work on LLM model testing, evaluation, and UI/UX refinement from a real user POV
• Access to pre-release systems for behavioral analysis
• Collaborations with teams who want high-quality failure data from a real-world tester, not sanitized corporate feedback