Pramodh R.

Key Skills

Software

Don't disclose

Other

Internal/Proprietary Tooling

Top Subject Matter

Large Language Models (LLMs) fine-tuning

Evaluation Domain Expertise

Rlhf Domain Expertise

Top Data Types

Text

Image

Audio

Top Task Types

Fine-tuning

Segmentation

Classification

Question Answering

RLHF

Prompt + Response Writing (SFT)

Computer Programming/Coding

Freelancer Overview

- Delivery Data Analyst at Turing with experience in AI training, fine-tuning, and evaluation workflows involving LLMs, SFT, and RLHF methodologies. - I have experience working on multiple AI training data and benchmarking projects, including Verification Benchmark QC, QC EXA AI Data Generation, Phoenix Trace Review QC, and Public Benchmark Data QC. - I primarily contributed as a rater, evaluating AI outputs for accuracy, consistency, reasoning quality, and guideline adherence across different workflows. I also supported projects as a reviewer during high-volume delivery periods to help maintain quality and delivery timelines.

Labeling Experience

Delivery Data Analyst | Turing — AI training/fine-tuning and evaluation work (LLMs, SFT, RLHF).

Don't discloseTextFine-tuning

Worked on fine-tuning large language models to improve performance and capabilities using supervised fine-tuning methods. Conducted side-by-side evaluations and RLHF workflows to compare model behavior and align outputs with desired quality. Enhanced multi-modality capabilities and math-reasoning performance so models can process and generate across diverse formats and tasks. • Applied Supervised Fine-Tuning (SFT) to optimize accuracy and functionality. • Performed SxS evaluations and RLHF for comprehensive model comparison. • Developed enhancements for multi-modal processing and generation. • Built improvements for math reasoning tasks and computational accuracy.

2024 - Present

AI Engineering Intern | Vyza Solutions — ML model training/deployment for financial back-testing.

OtherTextFine-tuning

Led end-to-end development and deployment of machine learning models for financial data back-testing, focusing on improving predictive accuracy for trading and portfolio decisions. Built an ensemble model (Random Forest) and integrated an ML backend into a full-stack web application for real-time analytics. The work involved preparing and using structured datasets for model training and evaluation to support forecasting performance in volatile market environments. • Developed and deployed ML pipelines for financial back-testing and optimization of trading strategies. • Engineered a Random Forest-based ensemble model and improved predictive accuracy by 15%. • Integrated the ML backend into a web application for real-time analysis and decisions. • Focused on robustness improvements for volatile market conditions.

2023 - 2024

Education

Work History