Independent AI Orchestrator & Technical Protocol Developer (Self-Directed / Technical Projects)
Executed RLHF-style alignment and evaluation by designing prompt sequences and running structured assessments on LLM outputs using low-level verification methods. Performed red-teaming and adversarial checks to measure robustness, contextual reasoning, and safety compliance of multilingual models. Validated technical behavior by detecting edge cases, hallucinations, and logical flaws through automated workflow scripts. • Engineered and optimized prompt sequences for LLM output evaluation • Conducted adversarial red-teaming protocols for multilingual robustness testing • Audited program logic for syntactic correctness, memory safety, and alignment • Built scripts/workflows to validate algorithmic behavior and identify failures