TAJA.
Analytics Engineer & Data Engineer
I build data models, pipelines, and BI layers that turn messy operational data into systems people actually trust.
Warsaw, Poland
01 / About
About Me
Analytics Engineer with 3+ years building data models, pipelines, and BI layers at Just Join IT. I work daily with SQL, Snowflake, dbt, and PostgreSQL — designing analytical layers that power business decisions across Sales, Product, Marketing, and Customer Success. Recently expanding into data engineering: GitHub Actions pipelines, Supabase architecture, and API integrations. I care about clean data, systems that scale, and making complex things simple for the people who use them.
Skills
Languages
- PolishNative
- RussianNative
- EnglishFluent
- FrenchIntermediate
02 / Experience
Experience
Nov 2025 — Present
Analytics Engineer
Just Join IT
- —Led high-impact data projects at the intersection of data engineering and analytics — including a full account merger (JustJoin IT + RocketJobs.pl) and an end-to-end BI stack migration; details in Work Projects below
- —Built event analytics layer — unpacking raw PostgreSQL events, designing tracking logic, syncing to Snowflake for downstream reporting
- —Designed multi-layer data models in Snowflake using dbt (bronze/silver/gold), following Kimball dimensional modeling principles — fact and dimension tables powering cross-team reporting
- —Executed production dbt migration of core mapping key (company_id → organization_unit_id) — full dependency audit across repo, safe rollout across two PRs, historical backfill using last_value ignore nulls window functions
- —Reduced technical debt by refactoring duplicated logic across dbt models into shared intermediate (int_*) layer
- —Integrated and modeled data from multiple sources including CRM (Pipedrive), product analytics and internal operational datasets
Apr 2024 — Oct 2025
Business Data Analyst
Just Join IT
- —Contributed analytics work to a product launch that drove 45% of company annual revenue in its first full year; partnered with Sales pre-launch — analyzing customer needs alongside usage data to shape go-to-market decisions
- —Designed churn calculation logic for a non-SaaS, package-based business model — defined criteria from scratch for clients with active packages expiring by usage or time, where standard SaaS metrics didn't apply
- —Built Power BI dashboards for marketing, sales and operational reporting
- —Developed early dbt models in Snowflake contributing to the company's analytical data layer
- —Cohort analysis, funnel tracking and performance analysis to identify growth opportunities and improve campaign effectiveness
- —Defined consistent business metrics and reporting logic across Sales and Marketing teams
Jul 2022 — Mar 2024
Response Rate Specialist
Just Join IT
- —Built an internal data aggregation tool in Google Sheets consolidating large datasets from multiple job listing portals — designed filter-based interface enabling non-technical CS and Sales teams to independently extract statistics and insights without analyst support
- —Designed and implemented a Twitter-based engagement strategy that increased response rates by 20% and engagement by 35%
- —Analyzed campaign data and user engagement metrics to identify performance patterns and improve outreach effectiveness
- —Built performance tracking dashboards in Google Sheets enabling ongoing monitoring and optimization of campaign results
Education
In progress, exp. August 2026
Data Engineer Course
Infoshare Academy
In progress, exp. 2027
Management, Master's
SWPS University Warsaw
2020 — 2023
Management & Leadership, Bachelor's
SWPS University Warsaw
03 / Projects
Projects
Personal Projects
Squash Performance Analytics
End-to-end analytics pipeline built with DuckDB, SQL and Python. Layered data models (staging, marts) feed a win probability prediction model based on historical performance, recent form, and match metrics. Makefile orchestration for a modular SQL + Python workflow.
Cheese Analytics Pipeline
Data pipeline to collect and analyze product data using Python, SQL, and DuckDB. Web scraping with requests + BeautifulSoup extracts structured data from a real-world website. Includes data cleaning, standardization, feature engineering, and a similarity scoring model for product recommendations. Modular Python scripts + Makefile orchestration.
Work Projects
Internal tools built at Just Join IT. Code is private — descriptions cover architecture and approach.
Internal Analytics Dashboards
Just Join IT, 2025–2026Context
Sales and Customer Success teams were managing client data in formula-heavy Google Sheets that broke at scale — slow, unsortable, no audit trail.
What I built
Two production internal dashboards replacing the legacy workflow entirely.
- —Designed full data architecture: Snowflake (dbt models, bronze/silver/gold) → Supabase (PostgreSQL + RLS + Auth) → Lovable (React frontend)
- —Automated daily Snowflake → Supabase sync via GitHub Actions (drop-recreate strategy with automatic schema drift propagation)
- —Implemented row-level security, audit logging, and allowlist-based RPC protection
- —Built virtualized UI handling ~26K rows with client-side sorting, multi-level filtering, and bulk edit with CSV import
- —Integrated webhook pipeline: Supabase → Make → Google Sheets for real-time audit trail
Outcome
Replaced manual spreadsheet workflows for two teams; priorities and changes fully auditable per user.
Code private — internal tooling
Account Merger — Data Migration
Just Join IT, 2025–2026Context
JustJoin IT and RocketJobs.pl are two brands under one company — and the time came to merge them into a single recruiter experience. Both platforms had separate employer accounts, separate logins, separate data. Merging incorrectly would mean one employer seeing another's data — a serious legal and trust issue.
What I did
- —Audited and deduplicated ~50K employer panels and 60K non-unique users → 24K panels, 40K unique users
- —Designed deduplication logic to identify true unique entities across two systems with different data models
- —Ensured zero data leakage between legally separate entities throughout the migration
- —Coordinated with CS and Sales teams for edge cases where data signals were ambiguous — delegating manual verification where data alone wasn't enough
Outcome
Clean, merged account structure with full data isolation — migration completed without incidents.
Code private — production data migration
04 / Contact
Let's Talk
Open to opportunities in analytics and data engineering. Reach out directly or find me on the usual places.
hiiii.taja@gmail.com