Abhi S.

Key Skills

Software

AWS SageMaker

OpenCV AI Kit (OAK)

Appen

Clickworker

Dataloop

Top Subject Matter

Medical imaging (DICOM/PACS) clinical labeling

Fintech risk explanation/decision support via retrieval-augmented generation

Legal Services & Contract Review

Top Data Types

Medical Dicom

Text

Geospatial Tiled Imagery

Top Task Types

Segmentation

Bounding Box

Point/Key Point

Text Generation

Text Summarization

RLHF

Fine-tuning

Data Collection

Transcription

Classification

Object Detection

Red Teaming

Evaluation/Rating

Polygon

Function Calling

Computer Programming/Coding

Freelancer Overview

AI Engineer — PaxeraHealth (medical imaging AI labeling/annotation & HIL review). Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. Education includes Bachelor of Science, Rutgers University. AI-training focus includes data types such as Medical, DICOM, and Computer Code and labeling workflows including Segmentation and Function Calling.

Labeling Experience

AI Engineer — PaxeraHealth (medical imaging AI labeling/annotation & HIL review)

Segmentation

Owned medical imaging data labeling and annotation workflows for DICOM studies, lesion candidates, segmentation masks, finding categories, and physician review outcomes. Defined labeling schemas, QA checks, review statuses, and curated labeled datasets for PyTorch/MONAI model training and evaluation. Supported human-in-the-loop review where users inspected findings, adjusted annotations, validated masks, and approved or rejected predictions for structured reporting. • Labeled DICOM-derived lesion candidates and segmentation masks with structured categories • Implemented labeling QA and review-state management to ensure dataset consistency • Prepared curated datasets for downstream training/evaluation in PyTorch/MONAI • Enabled clinician review UI interactions to refine annotations and validate outputs

2024 - Present

AI Engineer — Cross River (AI training/grounding via reviewed, structured outputs)

Function Calling

Built retrieval-based AI workflows that generate context-aware insights by assembling grounded inputs from structured data and operational records. Developed agentic AI workflow components that incorporate retrieval tools, validation checkpoints, and human-in-the-loop review patterns. Implemented production-style RAG/GraphRAG pipelines that rely on curated inputs and reviewable outputs for risk explanation and decision support. • Designed retrieval + context assembly pipelines for grounded outputs from domain records • Implemented agentic orchestration with validation steps for reviewable generation • Integrated model outputs into decision-support workflows with human-in-the-loop patterns • Supported structured response generation for internal AI/analytics applications

2020 - 2024

Education

Work History