HomeCompaniesChunkr

Open source API service to parse complex documents

Battle-tested + highly modular vision infrastructure to convert PDFs, PPTs, Word, Excel, PNG, and JPEGs into LLM-ready data. We started by building lumina.sh - where we needed to parse ~600M pages of scientific literature. The researchers didn't care - but devs wanted our ingestion pipeline. So we built chunkr instead. We offer high quality layout analysis, OCR, bounding boxes, granular VLM controls, semantic chunking, and all the last mile engineering that goes into building standout AI applications. Common use-cases include RAG, and automating document workflows like invoices/medical reports -> database.
Chunkr
Founded:2023
Team Size:3
Status:
Active
Location:San Francisco
Group Partner:Harj Taggar
Active Founders

Mehul Chadda, Founder

Co-founder & CEO at Chunkr. I have a background in metrology and a bsc in materials engineering. I work on helping computers read documents now.
Mehul Chadda
Mehul Chadda
Chunkr

Ishaan Kapoor, Founder

co-founder @ chunkr
Ishaan Kapoor
Ishaan Kapoor
Chunkr

Akhilesh Sharma, Founder

Co-Founder @ Lumina I am a mechanical engineer from the University of Illinois Urbana Champaign. I have experience in robotics and as a cloud solutions architect.
Akhilesh Sharma
Akhilesh Sharma
Chunkr
Company Launches
⚛️ Lumina - Accelerating Research with AI 🚀
See original launch post ›

👋 Hi everyone, we’re Mehul, Ishaan & Akhilesh - the founding team behind Lumina. Our research suite leverages LLM’s to help >8000 researchers discover, validate and curate scientific literature.

Mehul is a Materials Scientist and worked in Atom Probing R&D, Akhilesh is a national robotics champion and has experience as a Cloud Solutions Architect, and Ishaan studied AI and worked as a Data Engineer.

The Problem: Why does research take so long?

Science is slow. Most research, whether academic or in industry, takes months or years to do. Researchers have to dig through 100s, if not 1000s of scientific papers during this process, and have to vet and extract key insights that are relevant to their project.

As of now, a majority of the community depends on basic key-word search and manual skimming processes to do so - while wrestling paywalls. This can take anywhere from weeks to years.

💡 The Solution: An AI-powered Research Suite

We want to give researchers more time to do what's important. Lumina helps them cut the discovery and validation process down to minutes.

https://youtu.be/uX-Y7_LcWhI?si=LwZ8JgcExuwcjz4V

Some key features are:

  • 📚 Mini Lit-Reviews: Our answers find the most relevant sources for your query, and cite the key sections used from each journal article.
  • 📊 Tables: Extract key information from every source from the search - create a custom query, or use a preset to speed things up.
  • 🤖 Agents: Query multiple papers at once - connect our library and your collection for complex research tasks.

🌟 Vision: Speed up the entire scientific process

We will be augmenting the entire scientific process with AI. We’re building tools that will speed up data analysis, simulate experiments to generate data, bring multimodality and even write entire systematic reviews in one-click. The goal is to create an AGI-like experience for scientific research.

🙏 Our Ask:

📢 Try Lumina and let us know your thoughts!

🤝 Connect us to academic and private research groups.

Here’s a Calendly link for a short 15 min convo:

📅 https://calendly.com/mehul-yem/lumina-experience-interview

📧 You can reach us at mehul@lumina-chat.com

We’re also providing a 50% discount for Lumina Plus to celebrate our launch! Simply use this code on checkout: LAUNCHYC24

Company Photo

Company photo