Omar Alhait is a software engineer whose work spans a flyer-to-calendar Android application and document processing for a searchable public archive. A founding engineer at Reducto, he helped process the Epstein files for Jmail and demonstrated the extraction pipeline behind JFlights, where handwritten flight logs had to become structured records before an interface could make them searchable.
From photographed flyers to document processing
Omar graduated from Carnegie Mellon University in 2019 with a bachelor’s degree in electrical and computer engineering. His earlier project PictoCal, developed in 2015, addressed a familiar inconvenience: entering an event from a printed flyer into a calendar. The Android application let users photograph a flyer, review recognized words and dates, and approve a Google Calendar entry. Recognition handled the initial extraction; users could correct the result before creating the event.
His career subsequently included engineering at startups and venture-capital firms and bootstrapping a consumer-application company before joining Reducto. Reducto’s document-processing tools address files where embedded text alone cannot capture the information an application needs: scans, handwriting, complex layouts, and other visually structured material. Parsing recovers document content and structure; extraction organizes selected information into fields that software can use.
Turning difficult documents into usable records
Omar helped Reducto support the independent team building Jmail. Reducto processed documents and extracted email information as the project incorporated additional releases. Its contribution was the document-processing layer supporting the application, rather than authorship of the entire Jmail product.
In a joint presentation with Palak Agarwal, Omar demonstrated the extraction pipeline behind JFlights, an interface for exploring flight records. Handwritten logs illustrate the difficulty: useful records must be recovered from the page before they can be organized or searched. He showed how a schema specifies the desired fields, how developers inspect extraction results in Reducto Studio, and how a configured pipeline becomes available through an API. The schema connects the document’s irregular presentation to the consistent records the application expects.
Omar also discussed a consequential distinction between a PDF’s visible appearance and its underlying contents. Drawing a black box over text can leave that text accessible in the file’s metadata. He described how the team handled material intended to remain redacted—a case where extracting everything technically recoverable would conflict with preserving the intended redaction.
Palak Agarwal and Omar Alhait explain how Reducto turned difficult PDFs and handwritten flight logs into data for JMail—and why schemas, redactions, document-specific settings and representative evaluations matter before scaling.
JFlights needed a flight-specific extraction configuration that could handle varied handwritten and redacted documents under one schema.
▶ Watch ↗