Vansh P.

Key Skills

Software

Other

Clickworker

Top Subject Matter

AI agent orchestration and prompt-based automation for cybersecurity/research tasks

Machine learning experiment design

dataset preparation

Top Data Types

Text

Image

Video

Top Task Types

Prompt + Response Writing (SFT)

Text Generation

Question Answering

Fine-tuning

Data Collection

Computer Programming/Coding

Text Summarization

Evaluation/Rating

Object Detection

Freelancer Overview

Personal AI Operating System (Valianth): AI-driven multi-agent system (solo developer). Core strengths include Internal, Proprietary Tooling, and Other. Education includes Bachelor of Engineering with Honors, University of Bolton - Academic Centre RAK (2025) and Pearson BTEC Level 4 and Level 5 Diploma in Information Technology and Computing, UKCBC - UK College of Business and Computing (DIAC) (2024). AI-training focus includes data types such as Computer Code and Programming and labeling workflows including Prompt + Response Writing (SFT), Evaluation, and Rating.

Labeling Experience

AR Navigation + QR Bus Attendance (projects)

Other

Developed additional AI-adjacent prototypes that integrate with device inputs and user-facing interfaces, supporting practical testing and iterative refinement. The AR navigation prototype and QR attendance workflow focus on implementation, validation, and usability iteration rather than explicit human labeling. These projects support data handling and interaction logic required for downstream AI use cases. • Implemented an AR navigation prototype with iterative UI improvements • Built a companion website for navigation information and routes • Implemented QR attendance scanning/check-in with validation • Used Agile practices with user stories and sprint-based iteration

2026 - Present

AI Dice Prediction System (learning project)

Other

Created an AI/ML dice prediction prototype as a learning project focusing on end-to-end experimentation and model evaluation. Prepared and cleaned datasets, engineered basic features, and trained baseline models to compare performance across runs. Documented the approach, setup, and results for reproducibility. • Cleaned datasets and prepared samples for training • Engineered basic features and trained baseline models • Performed model evaluation and compared performance • Documented setup, results, and published to GitHub

2026 - Present

Personal AI Operating System (Valianth): AI-driven multi-agent system (solo developer)

Prompt + Response Writing (SFT)

Built a personal AI-driven multi-agent system intended to automate workflows via AI decision processes, requiring structured prompting and iterative improvements. Developed an agent architecture with role handoffs and guardrails, then integrated AI decision workflows with API-based actions. The work emphasizes evaluation via logging and iteration rather than manual dataset labeling. • Designed modular multi-agent roles, handoffs, and guardrails • Implemented state-machine logic for coordinating agent behaviors • Integrated structured prompting with API-based actions and logging • Iterated on prompts and workflow control using experiment feedback

2026 - Present

Education