The latest articles on DEV Community by Igor Nikolic (@igor_nikolic). https://dev.to/igor_nikolic https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3862558%2F7c29c536-a89d-4070-aa17-45f422e26e10.png https://dev.to/igor_nikolic en Igor Nikolic Sun, 05 Apr 2026 17:13:28 +0000 https://dev.to/igor_nikolic/filefeed-stop-writing-csv-parsers-for-every-new-client-1c34 https://dev.to/igor_nikolic/filefeed-stop-writing-csv-parsers-for-every-new-client-1c34 <h1> FileFeed: Stop Writing CSV Parsers for Every New Client </h1> <p>If you're building B2B SaaS, you know the pain. Every enterprise client sends data in their own format.</p> <p>One sends <code>Worker_ID, Name</code>. Another sends <code>EmpID, FullName</code>. A third sends <code>Mitarb_Nr, Naam</code>.</p> <p>You end up writing custom parsing code for each one. It works until you have 20 clients and 20 brittle scripts.</p> <p>We built <a href="https://filefeed.co" rel="noopener noreferrer">FileFeed</a> to fix this.</p> <h2> What is FileFeed? </h2> <p>FileFeed is a file processing platform for automated file ingestion. Files come in through SFTP, email, API, or cloud storage. FileFeed validates them, maps the columns to your schema, runs transformations, and delivers clean JSON to your product.</p> <p>One schema. Every client mapped to it. Zero custom parsers.</p> <h2> How it works </h2> <ol> <li>Your client drops a file to SFTP</li> <li>FileFeed detects the pipeline (client + folder match)</li> <li>Schema validation runs</li> <li>Field mappings applied</li> <li>Transformations executed</li> <li>Webhook fires to your system</li> <li>You fetch clean JSON via API</li> </ol> <p>Each client gets their own SFTP credentials, isolated storage, and a pipeline with their specific field mappings. You define the schema once. FileFeed normalizes everything to it.</p> <h2> What you get out of the box </h2> <p><strong>Multi-channel ingest</strong> : SFTP, email, API, cloud storage</p> <p><strong>Schema validation</strong> : reject bad data before it hits your system</p> <p><strong>Field mapping</strong> : <code>Worker_ID</code> to <code>employee_id</code>, per client, no code</p> <p><strong>Built-in transformations</strong> : toLowerCase, trim, formatPhoneNumber, and more</p> <p><strong>Webhook delivery</strong> : HMAC signed, fires on file received, processed, or failed</p> <p><strong>Outbound delivery</strong> : push clean data back via API, SFTP, cloud storage, or email</p> <p><strong>Dashboard</strong> : see every pipeline run, download original and processed files</p> <h2> What it replaces </h2> <p>Before FileFeed, teams used tools like Couchdrop for SFTP file routing. That handles file transfer but not file processing. Engineers still wrote all the parsing, mapping, and validation code themselves.</p> <p>FileFeed handles the full pipeline from ingest to delivery.</p> <h2> Who is this for? </h2> <p>B2B SaaS companies whose enterprise clients send data files. HR-tech, fintech, healthcare, supply chain. If your engineers are writing file parsers instead of building product, FileFeed handles that layer.</p> <h2> Try it </h2> <p>Check it out at <a href="https://filefeed.co" rel="noopener noreferrer">filefeed.co</a>.</p> automation data saas showdev