Sarthak P.

Key Skills

Software

Other

AWS SageMaker

Top Subject Matter

LLM training & evaluation for code generation (RLHF-style)

ML training pipelines and model evaluation/deployment (cloud-native)

Top Data Types

Text

Image

Top Task Types

RLHF

Fine-tuning

Object Detection

Tracking

Entity (NER) Classification

Freelancer Overview

Software Engineer — AI Training & Evaluation (Outlier AI | Remote). Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other and AWS SageMaker. Education includes Bachelor of Technology, Sri Siddhartha Institute of Technology, Tumkur (2027). AI-training focus includes data types such as Computer Code, Programming, and Text and labeling workflows including RLHF, Fine-tuning, and Object Detection.

Labeling Experience

AI/ML Engineer — Cloud Infrastructure & ML Pipelines (Freelance/Project-based | Remote)

AWS SageMakerFine-tuning

Designed and deployed end-to-end ML pipelines for model training, validation, and deployment using AWS infrastructure. Implemented CI/CD automation to streamline training and release workflows, improving overall deployment velocity. Orchestrated containerized model-serving workloads on Kubernetes and integrated monitoring for real-time performance tracking and cost optimization. • Preprocessed and framed large datasets for ML applications with validation • Automated model training/validation/deployment workflow via CI/CD • Managed Kubernetes (EKS) for high-availability inference services • Integrated CloudWatch monitoring for anomalies and cost optimization

2024 - Present

Outlier AI | Remote

AWS SageMakerImageObject DetectionTracking

No description provided.

2025 - 2025

AI Training and Evaluation Software Engineer - Outlier AI

AWS SageMakerTextRLHF

Designed and evaluated high-quality training samples for frontier LLMs across multiple programming languages to improve production model performance. Applied RLHF-style feedback loops to refine and rank AI-generated code outputs for correctness, efficiency, and alignment with industry standards. Performed prompt sensitivity analysis and code correctness grading to benchmark LLM behavior on real-world developer scenarios for measurable metric improvements. • Designed training/evaluation tasks for LLMs in Python, JavaScript, and C++ • Implemented RLHF-style ranking and refinement feedback loops • Conducted prompt sensitivity analysis and correctness grading • Collaborated with cross-functional AI research teams to communicate findings

2025 - 2025

Software Engineer — AI Training & Evaluation (Outlier AI | Remote)

OtherRLHF

Built and evaluated high-quality training samples for frontier LLMs across multiple programming languages to improve reasoning and instruction-following performance. Applied RLHF-style feedback loops to rank and refine AI-generated code outputs for correctness, efficiency, and alignment with coding standards. Performed prompt sensitivity analysis and graded code correctness to benchmark LLM performance in real developer scenarios. • Designed evaluation/rubric-driven grading for code outputs • Ranked model outputs using feedback-loop methodology • Ran prompt sensitivity experiments to measure robustness • Documented findings with cross-functional AI research teams

2025 - 2025

Education

Work History