Prathamesh D.

Key Skills

Software

Google Cloud Vertex AI

Labelbox

Label Studio

Mindrift

Scale AI

SuperAnnotate

Toloka

Telus

Mercor

Top Subject Matter

Finance

GUI

Generalist

Top Data Types

Document

Text

Image

Top Task Types

Prompt + Response Writing (SFT)

Evaluation/Rating

Fine-tuning

RLHF

Text Summarization

Data Collection

Question Answering

Bounding Box

Freelancer Overview

• Developed Reinforcement Learning from Human Feedback (RLHF) pipelines for financial reports, significantly improving multi-page context retention in LLM and Agentic AI systems. • Analysed 100+ financial statements using custom-built financial prompts, extracting structured insights for model training and investment decision support. • Led a cross-functional team of CAs and CFAs, building a structured prompt library for valuation processes and financial reasoning workflows. • Designed and executed test parameters across multiple financial datasets to validate and benchmark model performance. • Developed cross-referencing financial logic and hedge-based scenario models to improve AI-generated financial output accuracy. • Collaborated with the Apple Intelligence team to develop and refine a reasoning model, contributing to enhanced AI cognition capabilities. • Built a GUI Agent for ServiceNow, automating operational workflows and improving resolution efficiency by 30%. • Contributed to invoicing reasoning pipelines in collaboration with Intuit's engineering team.

Labeling Experience

Reasoning

DocumentPrompt + Response Writing (SFT)

Palo Alto, USA (Remote) | September 2024 – Present • Developed Reinforcement Learning from Human Feedback (RLHF) pipelines for financial reports, significantly improving multi-page context retention in LLM and Agentic AI systems. • Analysed 100+ financial statements using custom-built financial prompts, extracting structured insights for model training and investment decision support. • Led a cross-functional team of CAs and CFAs, building a structured prompt library for valuation processes and financial reasoning workflows. • Designed and executed test parameters across multiple financial datasets to validate and benchmark model performance. • Developed cross-referencing financial logic and hedge-based scenario models to improve AI-generated financial output accuracy. • Collaborated with the Apple Intelligence team to develop and refine a reasoning model, contributing to enhanced AI cognition capabilities. • Built a GUI Agent for ServiceNow, automating operational workflows and improving resolution efficiency by 30%. • Contributed to invoicing reasoning pipelines in collaboration with Intuit's engineering team.

2024 - 2026

Data annotations and RHLF

ImageRLHF

Palo Alto, USA (Remote) | September 2024 – Present • Developed Reinforcement Learning from Human Feedback (RLHF) pipelines for financial reports, significantly improving multi-page context retention in LLM and Agentic AI systems. • Analysed 100+ financial statements using custom-built financial prompts, extracting structured insights for model training and investment decision support. • Led a cross-functional team of CAs and CFAs, building a structured prompt library for valuation processes and financial reasoning workflows. • Designed and executed test parameters across multiple financial datasets to validate and benchmark model performance. • Developed cross-referencing financial logic and hedge-based scenario models to improve AI-generated financial output accuracy. • Collaborated with the Apple Intelligence team to develop and refine a reasoning model, contributing to enhanced AI cognition capabilities. • Built a GUI Agent for ServiceNow, automating operational workflows and improving resolution efficiency by 30%. • Contributed to invoicing reasoning pipelines in collaboration with Intuit's engineering team.

2024 - 2026

Education

Work History