You can bookmark a job description (it will be parsed), then paste a question and it generates an answer based on your resume, the job description, and your previously given answers for similar questions in other applications. The generated answer can be refined through a follow-up chat and exported as a PDF. It also works as a simple job application tracker.

Saves me tons of time and effort every day!

  Location: Barcelona, Spain
  Remote: Yes
  Willing to relocate: No
  Technologies: Python, FastAPI, JS, React, Ruby on Rails, Docker, AWS, PostgreSQL, Redis, Playwright, LiveKit (for voice AI)
  Résumé/CV: https://github.com/erichi/resume/blob/main/resume-eric-usmanov.pdf
  Email: e.usmanov@gmail.com

  Location: Barcelona, Spain
  Remote: Yes
  Willing to relocate: No
  Technologies: Python, FastAPI, JS, React, Ruby on Rails, Docker, AWS, PostgreSQL, Redis, Playwright, LiveKit (for voice AI)
  Résumé/CV: https://github.com/erichi/resume/blob/main/resume-eric-usmanov.pdf
  Email: e.usmanov@gmail.com

Comments URL: https://news.ycombinator.com/item?id=45615800

Points: 3

# Comments: 1

https://elmo.dev - a tool that automatically builds a searchable knowledge base around your project based on your conversation with coding tools.

It works automatically and doesn't require your attention. To build KB it uses same tool as you (claude/codex/gemini) so it uses the same quota and you don't have to pay additionally for the AI running it.

The result is ./elmodocs directory in the root of your project. You can reference CLAUDE.md/AGENTS.md/GEMINI.md to this directory or directly include the whole directory or its parts into the coding context.

Are you more interested in live streaming or just having a place to keep your coding replays?

Comments URL: https://news.ycombinator.com/item?id=45189260

Points: 36

# Comments: 12

  Location: Barcelona, Spain
  Remote: Remote Only
  Willing to relocate: No
  Technologies: Python(FastAPI), Ruby(Rails), JS, Node, Docker
  Résumé/CV: https://www.linkedin.com/in/ericusmanov
  Email: e.usmanov@gmail.com

Here is the link to VS Code extension - https://marketplace.visualstudio.com/items?itemName=scrapeco...

I've built browser automations for more than 5 years, and the constant frustration was always the sheer friction involved in getting working code – especially when debugging in headless mode or connecting to remote browsers.

When I started using LLMs to generate automation code, I found myself stuck in a repetitive loop: navigate to the desired page state, copy-paste HTML into the AI chat, and ask it to generate code. The worst part is that there was no easy way to run that generated code without losing the page state, forcing me to restart the browser session constantly. This wasted large amounts of time and mental energy. I built ScrapeCopilot to make this workflow seamless.

How it works:

ScrapeCopilot combines the power of a Jupyter-style notebook with a live Puppeteer browser session and integrated AI.

- Live Interactive Development: When you create an automation notebook, it initiates a fresh Puppeteer browser session. The page object is exposed directly to your notebook cells, allowing you to run any Puppeteer code against the live browser state and see the results instantly.

- AI-Powered Assistance: It integrates with GitHub Copilot (via the @scrapecopilot chat participant). The AI automatically sees the current page HTML, allowing it to generate highly relevant Puppeteer code based on your instructions directly within the chat.

- LLM Code Export: Once you've developed your automation logic, you can easily export the final, complete Puppeteer script based on your instructions.

This tool saves me hours daily, but even more importantly, it improves the developer experience in browser automation which is frustrating area.

I believe ScrapeCopilot can complement existing browser automation tools and frameworks by providing an interactive AI-assisted development experience.

Current Status & Future Plans:

- The extension currently works within VS Code. It will work in Cursor, but without chat support initially. I'm actively working on integrating a backend server to enable full chat functionality with Cursor.

- Currently the key workflow assumes that you create a new browser automation step by step, using code cells. But in my work I spend half of the time fixing existing automations, so my focus now is trying to adapt extension for debugging and fixing existing code.

- Playwright support is also on the list.

Check out short videos: - Demo: Headless False - https://scrapecopilot.ai/assets/demo-headless-false-Dhc_jeNR... - Demo: Headless True - https://scrapecopilot.ai/assets/demo-headless-true-PRQndDxP....

I'd love to hear your thoughts, feedback, and any suggestions!

Comments URL: https://news.ycombinator.com/item?id=44072704

Points: 3

# Comments: 0