[{"data":1,"prerenderedAt":1014},["ShallowReactive",2],{"project-en-govt-analyser":3,"project-nav-en":285},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"slug":10,"year":11,"featured":12,"order":13,"technologies":14,"github":7,"demo":7,"cover":20,"coverWidth":21,"coverHeight":22,"placeholder":6,"role":23,"category":24,"coverAlt":25,"architecture":26,"galleryTitle":31,"galleryDescription":32,"screenshots":33,"body":63,"_type":279,"_id":280,"_source":281,"_file":282,"_stem":283,"_extension":284},"\u002Fprojects\u002Fen\u002Fgovt-analyser","en",false,"","Brazilian Public Spending Analyzer","Ement: an application to compare funding allocated through Brazilian congressional budget amendments, search public records and export data.","govt-analyser","2026",true,2,[15,16,17,18,19],"Nuxt","TypeScript","FastAPI","PostgreSQL","D3","\u002Fimages\u002Fement\u002Foverview.webp",1440,1261,"Full-stack development","Data & transparency","Ement overview with committed, verified and paid amounts, amendment year trends and leading spending areas.",[27,28,29,30],"CGU · Open CSV data","Python · PostgreSQL","FastAPI · Queries and aggregation","Nuxt · Vue · D3","From national totals to individual allocations","Real application screens using Portal da Transparência data, covering state comparisons, record exploration and context for interpreting the figures. The application interface is in Portuguese.",[34,40,45,51,57],{"src":35,"title":36,"description":37,"alt":38,"width":21,"height":39},"\u002Fimages\u002Fement\u002Fmap.webp","Distribution across Brazil","Map of Brazil's 26 states and the Federal District, with 2025 commitments, state rankings and a state filter.","Ement map of Brazil showing committed amounts and state rankings.",1128,{"src":41,"title":42,"description":43,"alt":44,"width":21,"height":39},"\u002Fimages\u002Fement\u002Fmap-per-capita.webp","Resources per resident","Committed amounts divided by the reference population from the 2022 Census. The map and ranking update together.","Map and ranking of committed amendment amounts per resident, using IBGE population data.",{"src":46,"title":47,"description":48,"alt":49,"width":21,"height":50},"\u002Fimages\u002Fement\u002Frecords.webp","Search and export","Combined year, state and spending area filters, sorting and pagination. CSV export includes every matching record.","Amendment records filtered to health spending in São Paulo in 2024, with authors, destinations, amounts and CSV export.",1331,{"src":52,"title":53,"description":54,"alt":55,"width":21,"height":56},"\u002Fimages\u002Fement\u002Frecord-detail.webp","Amendment details","Author, location, program and budget action, with spending stages and commitments carried over from previous years.","Allocation details showing the destination, committed, verified and paid amounts, and commitments carried over from previous years.",1100,{"src":58,"title":59,"description":60,"alt":61,"width":21,"height":62},"\u002Fimages\u002Fement\u002Fsources.webp","Sources and methodology","Dataset coverage, indicator definitions and links to official CGU and IBGE sources.","Data methodology page showing the source, coverage of 86,871 records and interpretation criteria.",1213,{"type":64,"children":65,"toc":271},"root",[66,75,81,86,92,97,102,108,124,129,135,230,236,255,260,266],{"type":67,"tag":68,"props":69,"children":71},"element","h2",{"id":70},"understanding-where-public-funds-go",[72],{"type":73,"value":74},"text","Understanding where public funds go",{"type":67,"tag":76,"props":77,"children":78},"p",{},[79],{"type":73,"value":80},"In Brazil, members of Congress can direct public funding through amendments to the federal budget. Comparing states or investigating an area such as healthcare goes beyond looking up a single allocation: it requires filtering thousands of records, adding up amounts and distinguishing spending commitments from verified expenses and payments.",{"type":67,"tag":76,"props":82,"children":83},{},[84],{"type":73,"value":85},"I built Ement to bring those queries into one application. Visitors can start with a national overview, narrow their search and inspect individual allocations while keeping the same filters as they move between views.",{"type":67,"tag":68,"props":87,"children":89},{"id":88},"from-importing-data-to-exploring-it",[90],{"type":73,"value":91},"From importing data to exploring it",{"type":67,"tag":76,"props":93,"children":94},{},[95],{"type":73,"value":96},"My work covers the data import, API queries and aggregations, and the interface for exploring the results. The overview shows committed, verified and paid amounts, grouped by amendment proposal year, along with the leading spending areas and recipient states.",{"type":67,"tag":76,"props":98,"children":99},{},[100],{"type":73,"value":101},"The map compares states by committed amounts, payments or commitments per resident. Users can search records by year, state, spending area and text, then sort and page through the results. Each record brings together its author, location, program, budget action and detailed amounts. Filters are stored in the URL and kept when navigating between views.",{"type":67,"tag":68,"props":103,"children":105},{"id":104},"architecture",[106],{"type":73,"value":107},"Architecture",{"type":67,"tag":76,"props":109,"children":110},{},[111,113,122],{"type":73,"value":112},"Python scripts import the amendment dataset from ",{"type":67,"tag":114,"props":115,"children":119},"a",{"href":116,"rel":117},"https:\u002F\u002Fportaldatransparencia.gov.br\u002Fdownload-de-dados\u002Femendas-parlamentares",[118],"nofollow",[120],{"type":73,"value":121},"Brazil's Transparency Portal",{"type":73,"value":123}," into PostgreSQL. The FastAPI backend uses asyncpg to query records and calculate aggregates on the server.",{"type":67,"tag":76,"props":125,"children":126},{},[127],{"type":73,"value":128},"The frontend uses Nuxt, Vue and TypeScript, with components for filters, indicators, charts and record details. D3 draws the map from a local geographic file covering Brazil's 26 states and the Federal District. Per-resident calculations use population figures from the 2022 Census by IBGE, Brazil's national statistics agency.",{"type":67,"tag":68,"props":130,"children":132},{"id":131},"decisions-that-support-meaningful-comparisons",[133],{"type":73,"value":134},"Decisions that support meaningful comparisons",{"type":67,"tag":136,"props":137,"children":138},"ul",{},[139,151,161,171,210,220],{"type":67,"tag":140,"props":141,"children":142},"li",{},[143,149],{"type":67,"tag":144,"props":145,"children":146},"strong",{},[147],{"type":73,"value":148},"The same selection in every view:",{"type":73,"value":150}," indicators, the map, records and CSV exports share query parameters. Exports include all matching records, not just the visible page.",{"type":67,"tag":140,"props":152,"children":153},{},[154,159],{"type":67,"tag":144,"props":155,"children":156},{},[157],{"type":73,"value":158},"User input in queries:",{"type":73,"value":160}," filter values are passed as SQL parameters, and sorting is restricted to predefined options. Neither accepts arbitrary input as part of the SQL statement.",{"type":67,"tag":140,"props":162,"children":163},{},[164,169],{"type":67,"tag":144,"props":165,"children":166},{},[167],{"type":73,"value":168},"Reading the results:",{"type":73,"value":170}," totals and grouped results are queried within the same transaction. CSV exports read records in batches rather than loading the entire export into memory.",{"type":67,"tag":140,"props":172,"children":173},{},[174,179,181,187,189,194,196,201,203,208],{"type":67,"tag":144,"props":175,"children":176},{},[177],{"type":73,"value":178},"Separate spending stages:",{"type":73,"value":180}," commitments (",{"type":67,"tag":182,"props":183,"children":184},"em",{},[185],{"type":73,"value":186},"empenhado",{"type":73,"value":188},"), expenses verified as payable (",{"type":67,"tag":182,"props":190,"children":191},{},[192],{"type":73,"value":193},"liquidado",{"type":73,"value":195},") and payments (",{"type":67,"tag":182,"props":197,"children":198},{},[199],{"type":73,"value":200},"pago",{"type":73,"value":202},") have distinct indicators. Payments against commitments carried over from previous fiscal years (",{"type":67,"tag":182,"props":204,"children":205},{},[206],{"type":73,"value":207},"restos a pagar",{"type":73,"value":209},") are also identified separately.",{"type":67,"tag":140,"props":211,"children":212},{},[213,218],{"type":67,"tag":144,"props":214,"children":215},{},[216],{"type":73,"value":217},"Geographic limits:",{"type":73,"value":219}," records without a state remain in overall totals but are not assigned to a state on the map. Per-resident figures show the population reference year so readers can interpret the comparison.",{"type":67,"tag":140,"props":221,"children":222},{},[223,228],{"type":67,"tag":144,"props":224,"children":225},{},[226],{"type":73,"value":227},"Keyboard and mobile access:",{"type":73,"value":229}," the layout adapts to smaller screens, the map supports keyboard selection, and the detail dialog supports focus navigation.",{"type":67,"tag":68,"props":231,"children":233},{"id":232},"dataset-coverage-and-validation",[234],{"type":73,"value":235},"Dataset coverage and validation",{"type":67,"tag":76,"props":237,"children":238},{},[239,241,246,248,253],{"type":73,"value":240},"The dataset shown here contains ",{"type":67,"tag":144,"props":242,"children":243},{},[244],{"type":73,"value":245},"86,871 records",{"type":73,"value":247},", with amendment proposal years from ",{"type":67,"tag":144,"props":249,"children":250},{},[251],{"type":73,"value":252},"2014 to 2025",{"type":73,"value":254},". The application lets visitors compare funding distribution, look up an allocation and export their selection.",{"type":67,"tag":76,"props":256,"children":257},{},[258],{"type":73,"value":259},"Validation compares API totals against the source CSV down to the cent and checks combined filters, pagination, exports and invalid queries. A methodology page explains coverage and indicators, with links to the sources. The screenshots show the running application using this dataset.",{"type":67,"tag":68,"props":261,"children":263},{"id":262},"technologies",[264],{"type":73,"value":265},"Technologies",{"type":67,"tag":76,"props":267,"children":268},{},[269],{"type":73,"value":270},"Python · FastAPI · asyncpg · PostgreSQL · Piccolo · Nuxt · Vue · TypeScript · Nuxt UI · D3",{"title":7,"searchDepth":13,"depth":13,"links":272},[273,274,275,276,277,278],{"id":70,"depth":13,"text":74},{"id":88,"depth":13,"text":91},{"id":104,"depth":13,"text":107},{"id":131,"depth":13,"text":134},{"id":232,"depth":13,"text":235},{"id":262,"depth":13,"text":265},"markdown","content:projects:en:govt-analyser.md","content","projects\u002Fen\u002Fgovt-analyser.md","projects\u002Fen\u002Fgovt-analyser","md",[286,492,658,783,905],{"_path":287,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":288,"description":289,"slug":290,"year":7,"role":23,"featured":12,"order":291,"technologies":292,"github":7,"demo":7,"cover":294,"coverAlt":295,"coverWidth":21,"coverHeight":296,"category":297,"placeholder":6,"architecture":298,"screenshots":303,"body":334,"_type":279,"_id":489,"_source":281,"_file":490,"_stem":491,"_extension":284},"\u002Fprojects\u002Fen\u002Ffinance-platform","Blanco Finanças","A client portal for tracking savings plans and payments, with an admin area for managing plans and cash flow.","finance-platform",1,[15,17,18,293],"SQLAlchemy","\u002Fimages\u002Fblanco-financas\u002Fclient-dashboard.webp","Blanco Finanças client dashboard showing the balance, active plans, outstanding installments and payment history.",1000,"Web application",[299,300,301,302],"Nuxt · Vue · TypeScript","FastAPI · Application services","Domain · Financial rules","SQLAlchemy · PostgreSQL",[304,309,314,319,324,329],{"src":305,"title":306,"description":307,"alt":308,"width":21,"height":296},"\u002Fimages\u002Fblanco-financas\u002Fclient-plans.webp","Savings and goals","Client plans with progress, balances and upcoming due dates, available in compact and detailed views.","Detailed view of client savings plans, showing the status and progress of each plan.",{"src":310,"title":311,"description":312,"alt":313,"width":21,"height":296},"\u002Fimages\u002Fblanco-financas\u002Fclient-finance.webp","Installments and payments","A finance workspace with outstanding installments, amounts and due dates, plus access to withdrawals and payment history.","Client finance page showing installments, amounts, due dates and payment status.",{"src":315,"title":316,"description":317,"alt":318,"width":21,"height":296},"\u002Fimages\u002Fblanco-financas\u002Fadmin-dashboard.webp","Administrative overview","Client and plan counts, inflows, outflows, the protection fund and pending withdrawals in one view.","Admin dashboard with financial indicators and a breakdown of clients by status.",{"src":320,"title":321,"description":322,"alt":323,"width":21,"height":296},"\u002Fimages\u002Fblanco-financas\u002Fadmin-finance.webp","Cash flow","Transactions filtered by date range, category and direction, with totals and a breakdown of individual entries.","Admin cash flow page with date and category filters, financial totals and a transaction table.",{"src":325,"title":326,"description":327,"alt":328,"width":21,"height":296},"\u002Fimages\u002Fblanco-financas\u002Fadmin-plans.webp","Plan management","An administrative catalog showing plan status, savings limits and supported durations.","Plan management page with active and inactive plan cards, amounts and durations.",{"src":330,"title":331,"description":332,"alt":333,"width":21,"height":296},"\u002Fimages\u002Fblanco-financas\u002Fadmin-reports.webp","Reports","Date selection and export options for clients, transactions, cash flow and yields.","Reports page with date filters and Excel, CSV and PDF export options.",{"type":64,"children":335,"toc":481},[336,342,347,353,358,372,377,383,388,393,399,451,457,462,467,472,476],{"type":67,"tag":68,"props":337,"children":339},{"id":338},"two-workflows-in-one-application",[340],{"type":73,"value":341},"Two workflows in one application",{"type":67,"tag":76,"props":343,"children":344},{},[345],{"type":73,"value":346},"Blanco Finanças clients need to check their savings plan balances, see upcoming installments and track payments. Administrators need a different view: clients, the plan catalog, incoming and outgoing funds, and pending withdrawals. I built both areas around a shared API, with separate permissions for each role.",{"type":67,"tag":68,"props":348,"children":350},{"id":349},"what-i-built",[351],{"type":73,"value":352},"What I built",{"type":67,"tag":76,"props":354,"children":355},{},[356],{"type":73,"value":357},"My work covers the Nuxt frontend and FastAPI backend, including financial rules and integrations. The client portal shows plan progress, installments and payment history, with access to withdrawals.",{"type":67,"tag":76,"props":359,"children":360},{},[361,363,370],{"type":73,"value":362},"When signing up for a plan, clients see a recommendation and a cost breakdown before activating it through ",{"type":67,"tag":114,"props":364,"children":367},{"href":365,"rel":366},"https:\u002F\u002Fwww.bcb.gov.br\u002Fen\u002Ffinancialstability\u002Fpix_en",[118],[368],{"type":73,"value":369},"Pix, Brazil's instant payment system",{"type":73,"value":371},". The backend integrates with Mercado Pago to track transaction states and receive payment confirmations.",{"type":67,"tag":76,"props":373,"children":374},{},[375],{"type":73,"value":376},"Administrators can review key figures, filter transactions by date and category, manage the plan catalog and access reports on clients, transactions, cash flow and returns.",{"type":67,"tag":68,"props":378,"children":380},{"id":379},"application-structure",[381],{"type":73,"value":382},"Application structure",{"type":67,"tag":76,"props":384,"children":385},{},[386],{"type":73,"value":387},"The frontend uses Nuxt, Vue and TypeScript with Nuxt UI components. Pages handle navigation, while composables centralize API access for the client and admin screens.",{"type":67,"tag":76,"props":389,"children":390},{},[391],{"type":73,"value":392},"FastAPI handles incoming requests, and application services coordinate use cases. Domain entities and services contain the financial rules, separate from adapters for external services. Data is stored in PostgreSQL through asynchronous SQLAlchemy, with Alembic managing migrations.",{"type":67,"tag":68,"props":394,"children":396},{"id":395},"technical-decisions",[397],{"type":73,"value":398},"Technical decisions",{"type":67,"tag":136,"props":400,"children":401},{},[402,421,431,441],{"type":67,"tag":140,"props":403,"children":404},{},[405,410,412,419],{"type":67,"tag":144,"props":406,"children":407},{},[408],{"type":73,"value":409},"Monetary calculations:",{"type":73,"value":411}," I use integer cents and ",{"type":67,"tag":413,"props":414,"children":416},"code",{"className":415},[],[417],{"type":73,"value":418},"Decimal",{"type":73,"value":420}," in the backend, with rounding defined in the monetary value object. This makes the representation and rounding of amounts explicit in the financial rules.",{"type":67,"tag":140,"props":422,"children":423},{},[424,429],{"type":67,"tag":144,"props":425,"children":426},{},[427],{"type":73,"value":428},"API permissions:",{"type":73,"value":430}," authentication uses JWT, and backend operations check access by role. Client and admin access is enforced beyond the screens each user can see.",{"type":67,"tag":140,"props":432,"children":433},{},[434,439],{"type":67,"tag":144,"props":435,"children":436},{},[437],{"type":73,"value":438},"Payment confirmations:",{"type":73,"value":440}," the backend tracks transaction states and handles Pix confirmations idempotently, so a repeated confirmation does not apply the same transaction again.",{"type":67,"tag":140,"props":442,"children":443},{},[444,449],{"type":67,"tag":144,"props":445,"children":446},{},[447],{"type":73,"value":448},"Integration boundaries:",{"type":73,"value":450}," adapters handle Mercado Pago, Brazil's Central Bank, email delivery and report exports, keeping those details out of the domain rules.",{"type":67,"tag":68,"props":452,"children":454},{"id":453},"what-the-application-supports",[455],{"type":73,"value":456},"What the application supports",{"type":67,"tag":76,"props":458,"children":459},{},[460],{"type":73,"value":461},"Clients can track their plans and payments while administrators review transactions and manage plans within the same system. Both areas are implemented and shown in the screenshots on this page.",{"type":67,"tag":76,"props":463,"children":464},{},[465],{"type":73,"value":466},"In operation, the application served more than 100 users, with three administrators working at the same time. Bringing these workflows together reduced spreadsheet work by around 30 hours per week and allowed the operation to grow.",{"type":67,"tag":76,"props":468,"children":469},{},[470],{"type":73,"value":471},"The screenshots were taken from a local instance using test accounts and existing transactions. Identifying details have been hidden.",{"type":67,"tag":68,"props":473,"children":474},{"id":262},[475],{"type":73,"value":265},{"type":67,"tag":76,"props":477,"children":478},{},[479],{"type":73,"value":480},"Nuxt · Vue · TypeScript · Nuxt UI · FastAPI · Python · PostgreSQL · SQLAlchemy · Alembic",{"title":7,"searchDepth":13,"depth":13,"links":482},[483,484,485,486,487,488],{"id":338,"depth":13,"text":341},{"id":349,"depth":13,"text":352},{"id":379,"depth":13,"text":382},{"id":395,"depth":13,"text":398},{"id":453,"depth":13,"text":456},{"id":262,"depth":13,"text":265},"content:projects:en:finance-platform.md","projects\u002Fen\u002Ffinance-platform.md","projects\u002Fen\u002Ffinance-platform",{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"slug":10,"year":11,"featured":12,"order":13,"technologies":493,"github":7,"demo":7,"cover":20,"coverWidth":21,"coverHeight":22,"placeholder":6,"role":23,"category":24,"coverAlt":25,"architecture":494,"galleryTitle":31,"galleryDescription":32,"screenshots":495,"body":501,"_type":279,"_id":280,"_source":281,"_file":282,"_stem":283,"_extension":284},[15,16,17,18,19],[27,28,29,30],[496,497,498,499,500],{"src":35,"title":36,"description":37,"alt":38,"width":21,"height":39},{"src":41,"title":42,"description":43,"alt":44,"width":21,"height":39},{"src":46,"title":47,"description":48,"alt":49,"width":21,"height":50},{"src":52,"title":53,"description":54,"alt":55,"width":21,"height":56},{"src":58,"title":59,"description":60,"alt":61,"width":21,"height":62},{"type":64,"children":502,"toc":650},[503,507,511,515,519,523,527,531,541,545,549,620,624,638,642,646],{"type":67,"tag":68,"props":504,"children":505},{"id":70},[506],{"type":73,"value":74},{"type":67,"tag":76,"props":508,"children":509},{},[510],{"type":73,"value":80},{"type":67,"tag":76,"props":512,"children":513},{},[514],{"type":73,"value":85},{"type":67,"tag":68,"props":516,"children":517},{"id":88},[518],{"type":73,"value":91},{"type":67,"tag":76,"props":520,"children":521},{},[522],{"type":73,"value":96},{"type":67,"tag":76,"props":524,"children":525},{},[526],{"type":73,"value":101},{"type":67,"tag":68,"props":528,"children":529},{"id":104},[530],{"type":73,"value":107},{"type":67,"tag":76,"props":532,"children":533},{},[534,535,540],{"type":73,"value":112},{"type":67,"tag":114,"props":536,"children":538},{"href":116,"rel":537},[118],[539],{"type":73,"value":121},{"type":73,"value":123},{"type":67,"tag":76,"props":542,"children":543},{},[544],{"type":73,"value":128},{"type":67,"tag":68,"props":546,"children":547},{"id":131},[548],{"type":73,"value":134},{"type":67,"tag":136,"props":550,"children":551},{},[552,560,568,576,604,612],{"type":67,"tag":140,"props":553,"children":554},{},[555,559],{"type":67,"tag":144,"props":556,"children":557},{},[558],{"type":73,"value":148},{"type":73,"value":150},{"type":67,"tag":140,"props":561,"children":562},{},[563,567],{"type":67,"tag":144,"props":564,"children":565},{},[566],{"type":73,"value":158},{"type":73,"value":160},{"type":67,"tag":140,"props":569,"children":570},{},[571,575],{"type":67,"tag":144,"props":572,"children":573},{},[574],{"type":73,"value":168},{"type":73,"value":170},{"type":67,"tag":140,"props":577,"children":578},{},[579,583,584,588,589,593,594,598,599,603],{"type":67,"tag":144,"props":580,"children":581},{},[582],{"type":73,"value":178},{"type":73,"value":180},{"type":67,"tag":182,"props":585,"children":586},{},[587],{"type":73,"value":186},{"type":73,"value":188},{"type":67,"tag":182,"props":590,"children":591},{},[592],{"type":73,"value":193},{"type":73,"value":195},{"type":67,"tag":182,"props":595,"children":596},{},[597],{"type":73,"value":200},{"type":73,"value":202},{"type":67,"tag":182,"props":600,"children":601},{},[602],{"type":73,"value":207},{"type":73,"value":209},{"type":67,"tag":140,"props":605,"children":606},{},[607,611],{"type":67,"tag":144,"props":608,"children":609},{},[610],{"type":73,"value":217},{"type":73,"value":219},{"type":67,"tag":140,"props":613,"children":614},{},[615,619],{"type":67,"tag":144,"props":616,"children":617},{},[618],{"type":73,"value":227},{"type":73,"value":229},{"type":67,"tag":68,"props":621,"children":622},{"id":232},[623],{"type":73,"value":235},{"type":67,"tag":76,"props":625,"children":626},{},[627,628,632,633,637],{"type":73,"value":240},{"type":67,"tag":144,"props":629,"children":630},{},[631],{"type":73,"value":245},{"type":73,"value":247},{"type":67,"tag":144,"props":634,"children":635},{},[636],{"type":73,"value":252},{"type":73,"value":254},{"type":67,"tag":76,"props":639,"children":640},{},[641],{"type":73,"value":259},{"type":67,"tag":68,"props":643,"children":644},{"id":262},[645],{"type":73,"value":265},{"type":67,"tag":76,"props":647,"children":648},{},[649],{"type":73,"value":270},{"title":7,"searchDepth":13,"depth":13,"links":651},[652,653,654,655,656,657],{"id":70,"depth":13,"text":74},{"id":88,"depth":13,"text":91},{"id":104,"depth":13,"text":107},{"id":131,"depth":13,"text":134},{"id":232,"depth":13,"text":235},{"id":262,"depth":13,"text":265},{"_path":659,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":660,"description":661,"slug":662,"year":7,"role":663,"featured":12,"order":664,"technologies":665,"category":671,"placeholder":6,"retrospective":12,"status":672,"architecture":673,"diagram":678,"body":693,"_type":279,"_id":780,"_source":281,"_file":781,"_stem":782,"_extension":284},"\u002Fprojects\u002Fen\u002Fresy-booking","Resy Reservations","An application that monitored restaurant availability in New York and made reservations matching the client’s preferences.","resy-booking","Full-stack development and deployment",3,[666,667,668,669,670],"Python","React","Selenium","Nginx","VPS","Web application & automation","Retired after a platform change",[674,675,676,677],"Preferences · React","Backend · Python","Browser · Selenium","Reservations · Resy",{"title":679,"steps":680},"From preferences to a reservation",[681,684,687,690],{"title":682,"description":683},"Preferences","Restaurant, party size, dates and times.",{"title":685,"description":686},"Monitoring","Continuous checks of the booking page.",{"title":688,"description":689},"Availability","A table within the selected windows.",{"title":691,"description":692},"Reservation","Booking through the signed-in browser.",{"type":64,"children":694,"toc":773},[695,701,706,711,717,722,727,733,738,743,748,754,759,764,768],{"type":67,"tag":68,"props":696,"children":698},{"id":697},"finding-a-table-before-the-slot-disappeared",[699],{"type":73,"value":700},"Finding a table before the slot disappeared",{"type":67,"tag":76,"props":702,"children":703},{},[704],{"type":73,"value":705},"The client lived in New York and wanted to book tables at restaurants where reservations filled as soon as they became available. Checking the pages manually meant being there at just the right moment.",{"type":67,"tag":76,"props":707,"children":708},{},[709],{"type":73,"value":710},"I built an application where he could choose a restaurant, party size and acceptable date and time windows. The system then monitored availability on Resy and made a reservation when it found a matching option.",{"type":67,"tag":68,"props":712,"children":714},{"id":713},"from-the-interface-to-the-server",[715],{"type":73,"value":716},"From the interface to the server",{"type":67,"tag":76,"props":718,"children":719},{},[720],{"type":73,"value":721},"I developed the React frontend, the Python backend and the browser automation with Selenium. I also deployed the application to a VPS and configured Nginx as its reverse proxy.",{"type":67,"tag":76,"props":723,"children":724},{},[725],{"type":73,"value":726},"Monitoring ran on the server. The interface collected the client’s preferences, while the backend coordinated page checks and the steps needed to complete a reservation.",{"type":67,"tag":68,"props":728,"children":730},{"id":729},"keeping-the-browser-running-and-signed-in",[731],{"type":73,"value":732},"Keeping the browser running and signed in",{"type":67,"tag":76,"props":734,"children":735},{},[736],{"type":73,"value":737},"Maintaining an authenticated session was one of the harder parts. The automation needed access to the client’s account to complete a booking, and I used session cookies to preserve that access between interactions. Selenium located page elements using XPath selectors.",{"type":67,"tag":76,"props":739,"children":740},{},[741],{"type":73,"value":742},"Another issue appeared on the VPS: in that environment, the site did not load correctly in a headless browser. I configured a virtual display on the server so the browser could run with a graphical interface without a physical monitor.",{"type":67,"tag":76,"props":744,"children":745},{},[746],{"type":73,"value":747},"These adjustments were part of operating the application. Finding a table also required keeping a browser that could navigate the booking flow.",{"type":67,"tag":68,"props":749,"children":751},{"id":750},"use-and-retirement",[752],{"type":73,"value":753},"Use and retirement",{"type":67,"tag":76,"props":755,"children":756},{},[757],{"type":73,"value":758},"The client used the bot at least once a week for much of a year.",{"type":67,"tag":76,"props":760,"children":761},{},[762],{"type":73,"value":763},"After Resy changed its interface, the automation needed an update. The client chose to stop using it. Because the integration relied on page elements, changes to the platform required revisiting the selectors and navigation flow.",{"type":67,"tag":68,"props":765,"children":766},{"id":262},[767],{"type":73,"value":265},{"type":67,"tag":76,"props":769,"children":770},{},[771],{"type":73,"value":772},"Python · React · Selenium · Nginx · VPS",{"title":7,"searchDepth":13,"depth":13,"links":774},[775,776,777,778,779],{"id":697,"depth":13,"text":700},{"id":713,"depth":13,"text":716},{"id":729,"depth":13,"text":732},{"id":750,"depth":13,"text":753},{"id":262,"depth":13,"text":265},"content:projects:en:resy-booking.md","projects\u002Fen\u002Fresy-booking.md","projects\u002Fen\u002Fresy-booking",{"_path":784,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":785,"description":786,"slug":787,"year":7,"role":788,"featured":6,"order":789,"technologies":790,"category":794,"placeholder":6,"retrospective":12,"architecture":795,"diagram":800,"body":815,"_type":279,"_id":902,"_source":281,"_file":903,"_stem":904,"_extension":284},"\u002Fprojects\u002Fen\u002Fsoundcloud-prospecting","Music Prospecting on SoundCloud","Automation that found artists through recent releases, organized their profiles and offered a music producer’s services through direct messages.","soundcloud-prospecting","Automation development",4,[666,18,791,792,793],"Playwright","Beautiful Soup","Windows Task Scheduler","Data collection & automation",[796,797,798,799],"Genre search · SoundCloud","Collection and selection · Python","Profiles and contacts · PostgreSQL","Messaging · Playwright",{"title":801,"steps":802},"From a new track to outreach",[803,806,809,812],{"title":804,"description":805},"Discovery","Recent tracks in the producer’s genres.",{"title":807,"description":808},"Selection","Artist profiles and play counts.",{"title":810,"description":811},"Storage","Profile identifiers and contact status.",{"title":813,"description":814},"Message","A template with the artist and track names.",{"type":64,"children":816,"toc":895},[817,823,828,833,839,844,849,854,860,865,870,876,881,886,890],{"type":67,"tag":68,"props":818,"children":820},{"id":819},"finding-artists-to-work-with",[821],{"type":73,"value":822},"Finding artists to work with",{"type":67,"tag":76,"props":824,"children":825},{},[826],{"type":73,"value":827},"A music producer needed to find emerging artists in the genres he worked with. The research involved following new releases, visiting profiles and reaching out with an offer of production services and a free trial.",{"type":67,"tag":76,"props":829,"children":830},{},[831],{"type":73,"value":832},"I built a Python automation to run that search, organize the profiles it found and send messages through the producer’s account. The system ran on the client’s own computer, as he had requested.",{"type":67,"tag":68,"props":834,"children":836},{"id":835},"separate-discovery-and-contact-processes",[837],{"type":73,"value":838},"Separate discovery and contact processes",{"type":67,"tag":76,"props":840,"children":841},{},[842],{"type":73,"value":843},"Every 30 minutes, a script searched SoundCloud for the latest tracks in the selected genres. It then visited each artist’s profile and used play counts as the criterion for selecting artists early in their careers.",{"type":67,"tag":76,"props":845,"children":846},{},[847],{"type":73,"value":848},"Profiles were stored in PostgreSQL. Before adding an artist, the script checked the profile identifier against the database to avoid duplicate records. A flag tracked whether that artist had already been contacted.",{"type":67,"tag":76,"props":850,"children":851},{},[852],{"type":73,"value":853},"A second script retrieved profiles that had not yet been contacted and sent the messages. It randomly selected a template and filled in the artist’s SoundCloud name and the track that had led to their discovery. The message offered production services and a free trial. Sending followed variable intervals and a daily message limit.",{"type":67,"tag":68,"props":855,"children":857},{"id":856},"working-with-the-platform",[858],{"type":73,"value":859},"Working with the platform",{"type":67,"tag":76,"props":861,"children":862},{},[863],{"type":73,"value":864},"The main challenge was navigating SoundCloud and interacting through the producer’s signed-in account. I used Playwright, session tokens and Beautiful Soup for navigation and reading information from the pages.",{"type":67,"tag":76,"props":866,"children":867},{},[868],{"type":73,"value":869},"Windows Task Scheduler ran the routines on the client’s computer. Separate scripts allowed music discovery and artist outreach to run at different intervals while sharing records in PostgreSQL.",{"type":67,"tag":68,"props":871,"children":873},{"id":872},"use-by-the-producer",[874],{"type":73,"value":875},"Use by the producer",{"type":67,"tag":76,"props":877,"children":878},{},[879],{"type":73,"value":880},"The producer used the tool over an extended period. From what I observed, it found at least 10,000 profiles and sent thousands of messages. I no longer have the execution records to verify those figures, or data on how many contacts became customers.",{"type":67,"tag":76,"props":882,"children":883},{},[884],{"type":73,"value":885},"The outcome I can describe is automated research and outreach, supported by a profile database that tracked who had already received a message.",{"type":67,"tag":68,"props":887,"children":888},{"id":262},[889],{"type":73,"value":265},{"type":67,"tag":76,"props":891,"children":892},{},[893],{"type":73,"value":894},"Python · PostgreSQL · Playwright · Beautiful Soup · Windows Task Scheduler",{"title":7,"searchDepth":13,"depth":13,"links":896},[897,898,899,900,901],{"id":819,"depth":13,"text":822},{"id":835,"depth":13,"text":838},{"id":856,"depth":13,"text":859},{"id":872,"depth":13,"text":875},{"id":262,"depth":13,"text":265},"content:projects:en:soundcloud-prospecting.md","projects\u002Fen\u002Fsoundcloud-prospecting.md","projects\u002Fen\u002Fsoundcloud-prospecting",{"_path":906,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":907,"description":908,"slug":909,"year":7,"role":910,"featured":6,"order":911,"technologies":912,"category":915,"placeholder":6,"retrospective":12,"architecture":916,"diagram":921,"body":936,"_type":279,"_id":1011,"_source":281,"_file":1012,"_stem":1013,"_extension":284},"\u002Fprojects\u002Fen\u002Finvestment-tax-api","Investment Tax Calculation API","A Python API that accepted US stock market transactions in XLSX files and returned per-stock calculations and a consolidated total for a Ukrainian client.","investment-tax-api","API development and data processing",5,[666,913,914],"pandas","NumPy","API & data processing",[917,918,919,920],"Spreadsheet · XLSX","Reading and organization · pandas","Transaction matching · FIFO","Consolidated response · JSON",{"title":922,"steps":923},"From a spreadsheet to per-stock results",[924,927,930,933],{"title":925,"description":926},"Spreadsheet","Transactions in English or Ukrainian.",{"title":928,"description":929},"Organization","Purchases and sales grouped by stock.",{"title":931,"description":932},"Calculation","Matching against the earliest purchases.",{"title":934,"description":935},"Response","Per-stock values and a total in JSON.",{"type":64,"children":937,"toc":1005},[938,944,949,954,960,965,970,975,981,986,991,996,1000],{"type":67,"tag":68,"props":939,"children":941},{"id":940},"bringing-a-periods-transactions-together",[942],{"type":73,"value":943},"Bringing a period’s transactions together",{"type":67,"tag":76,"props":945,"children":946},{},[947],{"type":73,"value":948},"A Ukrainian client traded on the US stock market, mainly through swing trading, and needed to calculate the tax amounts associated with his transactions. The data came in an XLSX spreadsheet, usually covering a year.",{"type":67,"tag":76,"props":950,"children":951},{},[952],{"type":73,"value":953},"I developed the entire API in Python. It accepted the file, organized the transactions and returned JSON containing per-stock values and a consolidated total. I used pandas and NumPy for reading, processing and calculations.",{"type":67,"tag":68,"props":955,"children":957},{"id":956},"purchases-and-sales-of-the-same-stock",[958],{"type":73,"value":959},"Purchases and sales of the same stock",{"type":67,"tag":76,"props":961,"children":962},{},[963],{"type":73,"value":964},"The main challenge was each stock’s transaction history. A purchase could be sold in several parts, with further purchases and sales of the same stock throughout the period.",{"type":67,"tag":76,"props":966,"children":967},{},[968],{"type":73,"value":969},"I used a queue to match transactions: as sales were processed, they consumed the earliest purchases available in the file. This FIFO approach tracked the quantities associated with each transaction.",{"type":67,"tag":76,"props":971,"children":972},{},[973],{"type":73,"value":974},"Spreadsheet parsing was another concern. The client needed to submit documents in both English and Ukrainian, so the processing recognized the information in either language format.",{"type":67,"tag":68,"props":976,"children":978},{"id":977},"delivery-and-the-scope-of-this-account",[979],{"type":73,"value":980},"Delivery and the scope of this account",{"type":67,"tag":76,"props":982,"children":983},{},[984],{"type":73,"value":985},"The client used the API and confirmed that it worked for him. The delivery automated spreadsheet reading and calculation consolidation into a response another system could consume.",{"type":67,"tag":76,"props":987,"children":988},{},[989],{"type":73,"value":990},"I estimate that the automation saved roughly a month of manual work per year. This is a retrospective estimate, without tracked hours or confirmation from the client.",{"type":67,"tag":76,"props":992,"children":993},{},[994],{"type":73,"value":995},"The validation I received was the client’s feedback after using the API. This account does not include an independent tax review.",{"type":67,"tag":68,"props":997,"children":998},{"id":262},[999],{"type":73,"value":265},{"type":67,"tag":76,"props":1001,"children":1002},{},[1003],{"type":73,"value":1004},"Python · pandas · NumPy",{"title":7,"searchDepth":13,"depth":13,"links":1006},[1007,1008,1009,1010],{"id":940,"depth":13,"text":943},{"id":956,"depth":13,"text":959},{"id":977,"depth":13,"text":980},{"id":262,"depth":13,"text":265},"content:projects:en:investment-tax-api.md","projects\u002Fen\u002Finvestment-tax-api.md","projects\u002Fen\u002Finvestment-tax-api",1794934301468]