---
title: "Discussion on \"We Built a CV Scanner That Compares Skills Against Real Jobs. Then Production Exposed a Bottleneck.\" | Hashnode"
description: "Discussion on \"We Built a CV Scanner That Compares Skills Against Real Jobs. Then Production Exposed a Bottleneck.\". Learn how we diagnosed slow production CV scans, identified ONNX embedding bottlenecks, and used bounded caching to improve performance in Glassbox."
url: "https://hashnode.com/posts/we-built-a-cv-scanner-that-compares-skills-against-real-jobs-then-production-exposed-a-bottleneck/6aedcc3277f6db8a034bd049"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [We Built a CV Scanner That Compares Skills Against Real Jobs. Then Production Exposed a Bottleneck.](https://glassboxdata.hashnode.dev/we-built-a-cv-scanner-that-compares-skills-against-real-jobs-then-production-exposed-a-bottleneck?utm_source=hashnode&utm_medium=feed)

Learn how we diagnosed slow production CV scans, identified ONNX embedding bottlenecks, and used bounded caching to improve performance in Glassbox.

Nov 6 · 4 min read · glassboxdata.hashnode.dev  
Read the article

-   [#python](https://hashnode.com/tag/python)
-   [#machinelearning](https://hashnode.com/tag/machinelearning)
-   [#backend-development](https://hashnode.com/tag/backend-development)
-   [#performance](https://hashnode.com/tag/performance)

## Written by

[Glassbox](https://hashnode.com/@muturi)

@muturi · Nairobi, Kenya

  
[View profile](https://hashnode.com/@muturi)

## Responses

Markdown supported · ⌘↵ to post

No responses yet.
