Ajdin Salihović @dataminerinsights

Freelance

Have a data problem?
I build the fix.

I take projects end to end — scoping, backend and API development, containerised cloud deployment, and handover documentation your team can actually maintain. If it involves messy data, repetitive work, or documents nobody wants to read, it's probably my kind of project.

Tell me what you're building See my experience ↓

Services

What I build for clients.

service

LLM document automation

Invoices, contracts, forms, reports — extracted into structured, validated data with LLM pipelines, evaluation sets and human-auditable review workflows. The same discipline I apply at Hypatos, scaled to your documents.

  • llm apis
  • extraction schemas
  • rag
  • evaluation
service

Data pipelines & warehousing

Ingestion, ETL/ELT, modelling and orchestration on BigQuery, Snowflake, Redshift or Databricks — with the error handling, logging and validation that keep pipelines boring in the best way.

  • python
  • airflow
  • dbt
  • sql
  • gcp / aws
service

Web data acquisition & enrichment

Scraping, API integration, deduplication, enrichment and QA rules that turn the open web into structured records delivered to the tools your team already uses — Airtable, Notion, sheets, or a custom app.

  • scraping
  • third-party apis
  • llm agents
  • qa rules
service

Dashboards & automated reporting

Looker Studio dashboards and scheduled reporting built on clean, validated data — plus the automations that save your team a week every month.

  • looker studio
  • reporting
  • scheduling
  • integrations
everything else

Not on the list? Ask anyway.

If it's technical, I can build it or write it — scripts, integrations, internal tools, technical documentation, data cleanups, one-off automations, or the weird project that doesn't fit a category. Describe the problem; I'll tell you honestly whether I'm the right person for it, and what it would take.

Describe your problem →

Process

How working together goes.

  1. Scope

    A short call or email thread to understand the problem, the data, and what "done" looks like. Free, no commitment.

  2. Proposal

    A written plan: deliverables, timeline, and a fixed or milestone-based price. No surprises later.

  3. Build

    Iterative development with regular demos, so you see progress on real data — not just status updates.

  4. Handover

    Deployed, documented, and explained. You own the code, and your team knows how to run it.

Send an inquiry Email me directly

Work

Experience.

Every role below shaped how I build today: QA taught me to distrust data until it's validated, warehousing taught me scale, and LLM work taught me that context is the product.

AI Context Expert

Hypatos current
period Jul 2026 – present location Germany (remote)
  • Engineer the context layer behind LLM-based document processing — extraction schemas, prompts and ground-truth datasets that drive automated capture of structured data from unstructured business documents.
  • Build evaluation sets and error taxonomies from model output, feeding structured findings back to engineering to raise extraction accuracy and reduce manual review.
  • Automate data preparation, validation and reporting across large document corpora with Python and SQL.

python · sql · llm apis · prompt & context engineering · evaluation

Lead Data Acquisition & Automation Engineer

AQVC
period Apr 2026 – Jul 2026 location remote
  • Built automated acquisition and enrichment pipelines combining third-party APIs (Apollo, Airtable, Google Workspace) with LLM agents to research, structure and validate investor data at scale.
  • Designed reusable Python automation and multi-stage QA workflows that replaced manual research steps and enforced validation rules on every record.
  • Integrated LLM tooling into internal processes and documented repeatable workflows adopted by the wider team.

python · llm agents · apollo · airtable · google workspace apis

QA Engineer / SDET

ABC Tech Group
period Feb 2025 – Apr 2026 location Tuzla, BiH
  • Owned quality for a multi-platform banking product across web, iOS and Android — test design, regression, functional, UI and exploratory testing managed in TestRail.
  • Debugged issues across the stack using logs, API responses and multiple environments, working directly with developers and product.
  • Improved release confidence through systematic edge-case coverage and maintained test documentation.

test automation · testrail · api debugging · banking domain

Senior Data Engineer

Falcony AI
period Oct 2024 – Feb 2025 location remote
  • Designed and maintained BigQuery data warehouses and ETL/ELT pipelines on GCP supporting analytics and reporting for product and marketing teams.
  • Developed modular Python services for recurring ingestion, cleaning and transformation.
  • Integrated third-party APIs (Google Analytics, Hyros) for attribution data and delivered Looker Studio dashboards used across the organisation.

bigquery · gcp · python · looker studio · etl/elt

Senior Data Engineer

Quantox Technology
period Aug 2024 – Dec 2024 location remote
  • Spearheaded a custom data platform on PySpark and Databricks, transforming raw streaming data into gold-layer analytics datasets.
  • Wrote scalable, testable Python modules for validation, transformation and enrichment.

pyspark · databricks · streaming · medallion architecture

Data Engineer

Endava
period Oct 2022 – Jul 2023 location Sarajevo, BiH
  • Developed and maintained Python Airflow DAGs automating ETL processes, with advanced error-handling and logging.
  • Designed performant SQL queries and data models on Snowflake and Oracle for near real-time analytics use cases.

airflow · snowflake · oracle · python · sql

Data Engineer → Associate Data Engineer

Virgin Pulse
period Oct 2020 – Sep 2022 location Tuzla, BiH
  • Built and monitored production pipelines on Redshift and Snowflake using Python, Airflow and shell scripting.
  • Drove data quality and reliability improvements through validation checks, refactoring and clearer logging.

redshift · snowflake · airflow · shell · data quality

Software Developer

Cape Ann Enterprises
period Feb 2018 – Oct 2019 location BiH
  • Built and shipped web applications with Angular, Drupal, JavaScript, jQuery and HTML/CSS, with automated UI testing in Selenium.
  • Produced mockups and drafts for client sites and documented implementation step by step.

angular · drupal · javascript · selenium

Programming Instructor

Algorithmics Global & Coding Giants ongoing
period Feb 2022 – present location remote
  • Teach Python, C# and web fundamentals to students aged 6–18 — four years of practice turning complex technical ideas into clear, accessible explanations.

python · c# · web fundamentals · teaching

Earlier roles

2016 – 2020
  • Technical Support Analyst — Marvelsoft: log analysis, Sphinx documentation, GitLab.
  • Software Engineer — Alfa24: Vue.js, PostgreSQL.
  • Web Developer — Leftor d.o.o.: PHP, Yii, HTML5/CSS3.

Toolbox

Skills, grouped.

ai & llm
LLM-based document processing & data extraction, prompt and context engineering, RAG, embeddings & vector search, evaluation sets and error analysis, LLM/agent APIs, MCP tooling
backend
Python (FastAPI, Flask), REST API design & integration, background jobs and scheduling, asynchronous / event-driven processing, C#
frontend
JavaScript, Angular, Vue.js, HTML5, CSS3, responsive UI work — TypeScript / React actively learning
data & databases
Advanced SQL, PostgreSQL, Oracle, Snowflake, BigQuery, Redshift, MongoDB, Cassandra; DBT, Airflow, Fivetran, PySpark, Databricks; ETL/ELT, warehouse & lakehouse design
cloud & devops
Google Cloud Platform, AWS (S3, Athena), Azure, Docker, Kubernetes, Git/GitLab, Linux, logging & monitoring
quality
Test design & automation (Selenium, TestRail), data validation, debugging & performance analysis, technical documentation (Sphinx), Agile/Scrum, Jira