For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
V
Vedad Spahic

Vedad Spahic

AI Training Data Specialist | Applied AI Engineer

Bosnia Herzegovina flagSarajevo, Bosnia Herzegovina
$40.00/hrIntermediateOther

Key Skills

Software

Other

Top Subject Matter

Software / IT / SaaS
Cloud Infrastructure / Logs / Diagnostics
E-commerce / Product Catalog Data

Top Data Types

DocumentDocument
TextText
Computer Code ProgrammingComputer Code Programming

Top Task Types

ClassificationClassification
Evaluation/RatingEvaluation/Rating
Text SummarizationText Summarization

Freelancer Overview

I have hands-on experience building and validating AI training-data-adjacent workflows involving classification, structured extraction, evidence review, and quality control. At Bickles Hardware, I worked on AI-assisted product data extraction pipelines where I validated outputs against source pages, reviewed extracted attributes, corrected data-quality issues, and prepared clean structured datasets for downstream use. This included schema-aware extraction, consistency checks, error analysis, and improving output quality across dynamic product pages. At HYCU, I helped build an AI-assisted log-triage workflow that combined deterministic analysis with token-efficient, human-in-the-loop AI prompting for faster root-cause investigation. Across these projects, my strengths include classification-oriented review, document/text understanding, structured data validation, prompt-guided QA, and turning noisy raw inputs into reliable labeled or structured outputs for AI-related workflows.

IntermediateEnglishCroatianSerbianGermanSlovenian

Labeling Experience

Log Triage + Human-in-the-Loop AI Review

TextComputer Programming Coding
Helped build a workflow that prepares summarized, structured evidence from logs for efficient AI-assisted diagnosis. The work involved identifying relevant signals, structuring evidence, reducing noise, and improving consistency of diagnostic review.

Helped build a workflow that prepares summarized, structured evidence from logs for efficient AI-assisted diagnosis. The work involved identifying relevant signals, structuring evidence, reducing noise, and improving consistency of diagnostic review.

2026 - Present

AI-Assisted Product Data Extraction and Structured Evidence Pipeline (Bickles Hardware)

DocumentClassification
Designed and led an AI-assisted pipeline for extracting and validating structured product data from PDF orders and dynamic web pages. Responsible for orchestrating agentic coding tools and browser automation to capture accurate page-level data and evidence for quality assurance and downstream processing. Built robust CSV/JSON pipelines and integrated schema-aware validations for high-volume, audit-ready datasets. • Developed manifest-driven workflows for reproducible, resumable extraction tasks • Used Codex, Claude Code, Pinecone, and Chrome DevTools MCP/CDP as primary tools • Engineered agentic scrapers with prompt-engineered outputs and strong anti-bot resilience • Delivered thousands of accurately labeled documents and product entries across large-scale inventory

Designed and led an AI-assisted pipeline for extracting and validating structured product data from PDF orders and dynamic web pages. Responsible for orchestrating agentic coding tools and browser automation to capture accurate page-level data and evidence for quality assurance and downstream processing. Built robust CSV/JSON pipelines and integrated schema-aware validations for high-volume, audit-ready datasets. • Developed manifest-driven workflows for reproducible, resumable extraction tasks • Used Codex, Claude Code, Pinecone, and Chrome DevTools MCP/CDP as primary tools • Engineered agentic scrapers with prompt-engineered outputs and strong anti-bot resilience • Delivered thousands of accurately labeled documents and product entries across large-scale inventory

2024 - Present

AI-Assisted PDF Order Processing

DocumentEvaluation Rating
Worked on document-based AI matching workflows for PDF order processing, including item/vendor matching, structured interpretation of document content, and quality-focused review of output accuracy.

Worked on document-based AI matching workflows for PDF order processing, including item/vendor matching, structured interpretation of document content, and quality-focused review of output accuracy.

2025 - 2025

Education

U

University of Sarajevo

Bachelor of Science, Electrical Engineering

Bachelor of Science
Not specified
U

University of Sarajevo

Master of Science, Electrical Engineering

Master of Science
Not specified

Work History

B

Bickles Hardware

Software Engineer | Applied AI Engineer

Sarajevo
2024 - Present
H

HYCU

Software Developer

Sarajevo
2022 - 2024