Cosmo Prompt Collection V2
→
Summary
Designed and deployed an LLM prompt dataset pipeline; authored, evaluated, and expanded prompts for large-scale model training across multiple task domains.
Results-driven AI Implementation Specialist and Generalist Expert with over 4 years of hands-on experience building autonomous AI agents, deploying LLM-powered systems, and delivering high-quality human feedback for RLHF pipelines. Proven ability to integrate complex tool ecosystems and manage AI infrastructure on VPS environments, executing end-to-end AI workflows across diverse modalities. Combines deep technical proficiency in Python, JSON, YAML, and API integration with strong evaluation expertise, consistently maintaining annotation and evaluation accuracy above 98%.
Generalist Expert
N/A, N/A, N/A
→
Summary
Led nuanced quality assessment of AI-generated responses across diverse subject domains, applying LLM evaluation expertise to deliver high-quality human feedback for RLHF pipelines.
Highlights
Evaluated AI-generated responses across diverse subject domains, conducting nuanced quality assessments covering accuracy, helpfulness, reasoning, tone, and safety for leading AI research organizations.
Applied LLM evaluation expertise, prompt interpretation, and structured rationale writing to deliver high-quality human feedback for RLHF pipelines, ensuring model alignment and performance.
Demonstrated broad generalist knowledge across STEM, finance, healthcare, and real estate domains, effectively assessing complex AI model outputs and enhancing evaluation criteria.
Advanced AI Implementation Specialist & Model Trainer (RLHF)
N/A, N/A, N/A
→
Summary
Engineered and deployed autonomous AI agents, managed AI infrastructure on VPS environments, and integrated AI systems with client tools, overseeing a portfolio of 10+ concurrent projects.
Highlights
Built and deployed autonomous AI agents, including sales, CSM, content, and reporting dashboards, integrating them into client production environments for enhanced operational efficiency.
Set up and managed AI infrastructure on VPS environments (Hostinger, GCP), configuring servers, environments, and network settings to ensure reliable uptime and optimal performance.
Integrated AI systems with client tools (Slack, Google Sheets, Monday.com, CRM) via REST APIs and automation frameworks (n8n, Make, Zapier), streamlining workflows and improving data flow.
Managed a portfolio of 10+ concurrent client projects end-to-end, from onboarding through build and delivery to ongoing maintenance, consistently maintaining a sub-24-hour response SLA.
Produced comprehensive technical documentation for every deployed system, enabling full replication by any team member and ensuring knowledge transfer.
Led LLM evaluation and optimization using RLHF techniques, including ranking, scoring, structured feedback, hallucination detection, and edge case identification to improve model accuracy.
Conducted advanced prompt engineering, prompt testing, and response validation across diverse domains to improve model alignment and overall task performance.
Administrative Officer
N/A, N/A, N/A
→
Summary
Oversaw administrative operations, served as primary staff contact across facilities, supervised departmental workflows, and provided advisory support to senior management.
Highlights
Oversaw comprehensive administrative operations, serving as the primary contact for staff across multiple facilities regarding supplies, scheduling, and expenses.
Supervised departmental workflows and developed effective operating procedures, improving organizational efficiency and compliance.
Provided advisory support to senior management on critical personnel and policy matters, contributing to strategic decision-making and operational improvements.
Remote Customer Support Specialist
N/A, N/A, N/A
→
Summary
Handled inbound support tickets and live chats for a U.S.-based e-commerce platform, achieving a 95% customer satisfaction rating.
Highlights
Handled inbound support tickets and live chats for a U.S.-based e-commerce platform, consistently achieving a 95% customer satisfaction rating.
Resolved complex order issues, processed returns, and provided comprehensive product guidance across Zendesk, Slack, and Shopify in collaboration with international teams.
Customer Service Representative
N/A, N/A, N/A
→
Summary
Resolved 95%+ of customer issues on first interaction and led a process optimization initiative that reduced response time by 20%.
Highlights
Resolved over 95% of customer issues on first interaction, demonstrating exceptional problem-solving skills and efficiency.
Led a process optimization initiative that successfully reduced response time by 20%, significantly enhancing customer service delivery.
Managed global client inquiries via email, chat, and phone channels, consistently maintaining high service standards and client satisfaction.
Freelance AI Agent Developer & Video Evaluator
N/A, N/A, N/A
→
Summary
Designed and delivered custom AI automation workflows and agent systems, managing the full client lifecycle while reducing manual effort by 60-80% for diverse clients.
Highlights
Designed and delivered custom AI automation workflows and agent systems for clients across e-commerce, content, and operations verticals, enhancing their operational efficiency.
Integrated client tools via APIs and automation platforms (n8n, Make, Zapier) to build end-to-end workflows, reducing manual effort by 60-80% per client.
Managed the full client lifecycle independently, encompassing scoping, build, delivery, documentation, and post-launch support, consistently delivering projects on time.
Performed high-quality video annotation and evaluation in support of AI training workflows, maintaining strict dataset quality and consistency standards.
→
Bachelor of Science
Shipping Management
→
Higher National Diploma
Shipping & Port Management
Issued By
micro1
Issued By
Invisible Technologies
Issued By
Invisible Technologies
Reinforcement Learning from Human Feedback (RLHF), Large Language Models (LLM), Prompt Engineering, Prompt Evaluation, AI Model Evaluation, Multimodal AI (Text, Image, Audio, Video), Response Quality Assessment, Hallucination Detection, Edge Case Identification, Error Analysis, Reward Model Training.
Autonomous AI Agent Design & Deployment, LLM Integration & Deployment, Workflow Automation (n8n, Make/Integromat, Zapier), API Integration (Shopify, Slack, GoHighLevel, Monday.com, Google Sheets), LangChain (familiar), REST APIs.
Python, SQL, JSON, YAML, Bash Scripting, Google Workspace, Process Mapping, Data Analysis, DevOps Fundamentals.
VPS Setup & Management (Hostinger, GCP, DigitalOcean), Linux Server Configuration, Environment Management, LabelBox, CVAT, Feather, Gala Platform, Smartcat, Scale AI, Remotasks, UHRS.
Data Annotation (Text, Image, Audio, Video), Dataset Curation, Content Moderation, Rubric Writing, Quality Assurance, Quality Review, Audit Trail Documentation, Gold Standard Definition.
English Language Evaluation (C2), Finance Expert, Generalist Expert, Real Estate, Health Care Specialist, Analytical Thinking, Technical Documentation, Client Account Management, Remote Collaboration.
→
Summary
Designed and deployed an LLM prompt dataset pipeline; authored, evaluated, and expanded prompts for large-scale model training across multiple task domains.
→
Summary
Developed clear, well-structured rubrics and benchmark training data to evaluate AI models on professional-grade deliverables.
→
Summary
Led systematic error detection and correction across LLM response datasets within the RLHF pipeline, improving output quality consistency through structured feedback and detailed error categorization across large evaluation batches.
→
Summary
Rewrote and refined AI-generated responses to reflect preferred model behavior, supporting reward model training within the RLHF pipeline.
→
Summary
Built a video analysis and Q&A system integrating AI agents for automated content evaluation and reporting.
→
Summary
Led image annotation, bounding box labeling, and QA review workflows supporting multimodal model training.