Andrej F.

Key Skills

Software

AWS SageMaker

Scale AI

Internal/Proprietary Tooling

Top Subject Matter

Enterprise Conversational AI / Multimodal Orchestration

Retrieval-Augmented Generation / Knowledge Base AI

Computer Vision / Multimodal AI

Top Data Types

Document

Image

Text

Top Task Types

Transcription

Data Collection

Classification

Red Teaming

RLHF

Prompt + Response Writing (SFT)

Evaluation/Rating

Freelancer Overview

AI Platform Engineer - Audio Data Workflow and Labeling. Brings 19+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. Education includes Specialist in Radio and Nuclear Physics, National University of Ukraine (2001). AI-training focus includes data types such as Audio, Document, and Image and labeling workflows including Transcription and Data Collection.

Labeling Experience

AI Platform Engineer - Vision Data Annotation & Collection

ImageData Collection

Directed agent-driven vision workflows involving the capture, annotation, and analysis of images and videos for multimodal AI orchestration. Implemented guided capture flows and annotation of screenshots and short video segments to support local computer vision pipelines. Managed the secure transfer and processing of visual data within air-gapped, enterprise-grade AI systems. • Developed vision capture and annotation interfaces for structured data collection. • Enabled secure, on-premises labeling and transfer of sensitive visual data. • Integrated vision-based AI pipelines for image and video analysis using internal tools. • Oversaw enrichment of multimodal data streams for improved AI memory and reasoning.

2025 - Present

AI Platform Engineer - Document Labeling & Embedding

DocumentData Collection

Established a private document knowledge base using retrieval-augmented generation (RAG) by ingesting, embedding, and storing customer-uploaded documents. Supervised data ingestion, vector embedding, and annotation workflows for secure, on-premises document enrichment supporting enterprise chat applications. Facilitated searchable, privacy-focused document retrieval by orchestrating labeling, embedding, and memory enrichment processes for long-term AI learning. • Orchestrated local document embedding and vector management via Qdrant. • Ensured secure, private handling and enrichment of customer documents within air-gapped systems. • Supported context enrichment for AI conversations through document-based memory retrieval. • Designed robust, automated document labeling workflows integrated with LLM-driven processes.

2025 - Present

Education

Work History