Data Analyst

Built a Python data pipeline that collects job postings from five sources, including LinkedIn, Djinni, RemoteOK, and Greenhouse/Lever APIs. Implemented deduplication and maintained a continuously updated dataset of 148 postings, reducing collection time from approximately three hours of manual work to 15 minutes.

Developed a rule-based text analyzer to extract job roles, seniority levels, and required skills. Created reporting to rank frequent skill gaps across near-fit positions and integrated an LLM workflow for CV-to-posting fit scoring with a validation step for unsupported claims.

Built a Python pipeline to aggregate event listings from seven sources and merge duplicate listings into a single dataset of approximately 300 events. Used Playwright, BeautifulSoup, and pandas for collection and processing.

Analyzed around 100 personal event ratings by source, venue, theme, and price. Identified that price data was missing for about 95% of rated events, added sources that publish pricing, and increased dataset price coverage from roughly 9% to 35%; also reviewed legal notices before enabling collection.

Analyze NFT market data, including floor prices, trading volume, and liquidity, to evaluate potential entry and exit points for leading collections. Research DeFi protocols across Ethereum, Arbitrum, and Optimism networks.

Build Python scripts to collect and process on-chain and market data, reducing manual monitoring. Use SQL and Excel to track transactions, portfolio performance, and P&L across multiple wallets, and document findings in structured decision-support reports.

Supported a digital transformation initiative by applying data analysis to commercial and operational reporting. Transformed raw business data into strategic insights for performance optimization.

Automated seven recurring reports sourced from Excel, CRM, and ERP systems. Reduced manual reporting workload from approximately 80 hours to 20 hours, representing a 75% reduction in effort.

Maintained six to seven core Excel reporting tables used by multiple teams and replaced manual copying with SQL exports, templates, and formulas. Reduced weekly update time by 2.5 times, from approximately three to four hours to around 1.5 hours.

Calculated subscription KPIs by region using SQL, including content views, user base, average subscription price, customer lifetime, and churn. Evaluated advertising campaigns using Google Analytics, identified checkout payment issues and sales drops, and provided findings that supported campaigns, payment-provider changes, and issue resolution.

Analyzed traffic and advertising performance data for more than 10 publishers and ad networks using Google Analytics, Yandex Metrica, and LiveInternet. Investigated cross-platform discrepancies and sudden changes in impressions and clicks.

Screened newly onboarded websites and identified bot traffic on approximately 30–35% of them through page, time, and regional traffic-distribution analysis. Monitored live traffic in Kibana, investigated advertiser complaints, configured ad templates, and handled 20–30 or more technical requests per day.