About the role
About Fresco
Fresco is building the operating system for the $12T construction industry. We (Akhil and Arvind) started Fresco after watching general contractors lose millions on bad bids because estimators were manually reading 500-page blueprint PDFs and typing numbers into spreadsheets. We're replacing that entire workflow with AI.
We're growing 140% month over month and need help scaling.
The Technical Problem
Construction blueprints aren't clean data. They're enormous, spatially complex PDFs full of ambiguous annotations, inconsistent scales, and domain-specific shorthand that no off-the-shelf model handles well. You'll be building systems that:
- Parse and extract structured line items from messy, multi-hundred-page blueprint PDFs at scale, optimizing our Python-based extraction engine to handle GBs of spatial data with low latency
- Ship LLM-powered features that turn unstructured construction documents into accurate, bid-ready estimates
- Close the loop with users: talk to an estimator in the morning, ship a fix by afternoon
Your First 90 Days
- By Day 30: Own and ship improvements to all core parts of our product, targeting a 90% reduction in extraction inaccuracies where applicable.
- By Day 60: Identify and resolve the top three data-quality bottlenecks reported by users, with measurable accuracy improvements in production.
- By Day 90: Propose and prototype a major feature informed by direct user conversations, something that changes how estimators interact with the product daily.
Our Stack
React + TypeScript on the frontend. Python on the backend. We obsess over getting the data model right so the rest of the code flows easily. We create pull requests for visibility but skip code review. We optimize for shipping speed, not process.
