Top all-time · #2810
Nanonets Blog
Company blog · nanonets.com/blog · on HN as sarthakjainaidata
Nanonets Blog is the company blog at nanonets.com, writing about ai and data. Since 2018, 27 different Hacker News readers have linked it 94 times, most often “DeepSORT: Deep Learning to Track Custom Objects in a Video”. It is still writing: the last post was on 2026-07-01, and you can follow it by feed.
Too few sampled comments (19) to say how often HN recommends it.?
Why HN links it
Nanonets' company blog covers OCR, document data extraction and deep learning tutorials. HN commenters mostly link its posts as promotion or reference, such as receipt OCR, Raspberry Pi object detection and structured LLM outputs.
Summarised by Claude from a sample of 19 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Nanonets automates messy manual processes including accounts payable, claims processing, order management, and more, and delivers clean data to systems of record like SAP, Salesforce, and more. Trusted by 10,000+ enterprises.
Most-linked posts?
- /3 people · 4 mentions · first linked 2018-01-17
- DeepSORT: Deep Learning to Track Custom Objects in a Video2 people · 5 mentions · first linked 2019-07-23
- Extracting Information from Receipts with GCNs2 people · 4 mentions · first linked 2020-03-26
- How to build a handwriting reader from scratch with deep learning2 people · 3 mentions · first linked 2020-08-27
- Automating receipt processing with deep learning2 people · 3 mentions · first linked 2020-01-27
- /research/nanonets-ocr-2/2 people · 3 mentions · first linked 2025-10-14
- /drone2 people · 2 mentions · first linked 2018-06-12
- /cookbooks/structured-llm-outputs/2 people · 2 mentions · first linked 2026-01-21
- We Ran 16 AI Models on 9k Real Documents. Here's What We Found2 people · 2 mentions · first linked 2026-03-11
- Show HN: Open-source 3B param model better than Mistral OCR2 people · 2 mentions · first linked 2025-06-12
What Hacker News says?
Not trying to just put in a baseless plug, but most of what you say can be refuted if you try out our product. Go here: https://nanonets.com/ocr-api/
Comment linking /ocr-api/
on “Why is AI so useless for business?”Submitted Building a license plate reader from scratch with deep learning as a story · 230 points
HN discussionI am Prathamesh, co-founder of https://nanonets.com We were working on a project to detect objects using deep learning with raspberry pi and we have benchmarked various deep learning architectures on pi. With ~100-200 images, you can create a detector of your own with this method. In this post, we have detected vehicles in Indian traffic using pi and also added github links to code to train the model on your own dat…
Comment linking /
on “How to Easily Detect Objects with Deep Learning on Raspberry Pi”Submitted Table Detection and Extraction Using Deep Learning as a story · 154 points
HN discussionFor those who don't like to wait for access, https://nanonets.com Just upload your training data and we will provide machine learning API automatically.
Comment linking /
on “Cloud AutoML: Making AI accessible to every business”Submitted A guide to OCR with Tesseract, OpenCV and Python as a story · 130 points
HN discussionOf course you can train deep learning models on small datasets. Where did you hear otherwise? Try yourself at https://nanonets.com/
Comment linking /
on “Google’s self-training AI turns coders into machine-learning masters”Submitted Automating receipt processing with deep learning as a story · 129 points
HN discussionA routine automation task is information extraction from any kind of document with a visual layout - receipts, PDFs, web pages. [1] [2] [1] : https://nanonets.com/blog/information-extraction-graph-convo... [2] : https://arxiv.org/pdf/1903.11279.pdf
Comment linking /blog/information-extraction-graph-convolutional-networks/
on “Deep Graph Library: Easy Deep Learning on Graphs”Submitted Show HN: OCR Benchmark Focusing on Automation as a story · 58 points
HN discussionI used a video's frames of myself playing RDR2 to train a neural network. I then showed it a photo my son took of some cows at the in-law's place. Here's what happened: https://photos.app.goo.gl/sDTofashQdDSUJMdA The network was trained with https://nanonets.com , if anyone is interested. Interesting stuff, given the knowledge came from synthesized horses. I did not train or segment any other entities other than Art…
Comment linking /
on “How the West Was Digitized: The Making of Rockstar Games’ Red Dead Redemption 2”Submitted A guide to Semantic Segmentation as a story · 58 points
HN discussionHey MrTonyD. I understand where your coming from. But honestly most people are using this for monitoring things like Solar Panels, Wind Turbines, detecting seals and crop prediction. We can't have a sustainable future without a lot of these technologies in place. https://nanonets.com/drone/wind-turbines/ https://nanonets.com/drone/solar-panels/ All new technology will have some deviant use case associated with it. T…
Comment linking /drone/solar-panels/
on “How to Do Real Time Object Detection on Drones”Submitted DeepSORT: Deep Learning to Track Custom Objects in a Video as a story · 58 points
HN discussionMaybe this is helpful: https://nanonets.com/blog/receipt-ocr/ In my Opinion Tesseract is the most sophisticated "free" OCR solution out there. The problem with Tesseract is not its recognition capabilities, but more the preprocessing steps. - thresholding - deskewing - segmentation - ... There is a C# library (non-free), that improves recognition A LOT, just by providing these abilities: https://www.vintasoft.com/vs…
Comment linking /blog/receipt-ocr/
on “Ask HN: OCR Libraries for Receipt Scanning/Parsing?”Submitted How to build a handwriting reader from scratch with deep learning as a story · 50 points
HN discussionHey guys, here's the updated URL: https://nanonets.com/blog/handwritten-form-ocr-handwriting-r...
Comment linking /blog/handwritten-form-ocr-handwriting-recognition/
on “How to build a handwriting reader from scratch with deep learning”Submitted Deep learning behind document processing as a story · 49 points
HN discussionExcited to share Nanonets-OCR2, a state-of-the-art suite of models designed for advanced image-to-markdown conversion and Visual Question Answering (VQA). Live Demo -> https://docstrange.nanonets.com/ Blog -> https://nanonets.com/research/nanonets-ocr-2/
Comment linking /research/nanonets-ocr-2/
on “Nanonets-OCR2-3B – OCR model that transforms documents into structured markdown”Submitted ID Card Digitization and Information Extraction using Deep Learning as a story · 49 points
HN discussionHey guys, we are building https://nanonets.com/drone We were working with a customer from South Africa to monitor construction progress using drone imagery and wanted to share case study based on our experience here. Hope it helps!
Comment linking /drone
on “How to Apply Deep Learning to Drone Imagery”Submitted Table Extraction Using LLMs as a story · 18 points
HN discussionHey guys, we are building https://nanonets.com/drone . We were working with a customer from South Africa to monitor construction progress using drone imagery and wanted to share case study based on our experience here. Hope it helps!
Comment linking /drone
on “How to Easily Do Object Detection on Drone Imagery Using Deep Learning”Submitted The unreasonable effectiveness of language models in document processing as a story · 16 points
HN discussionResults with our benchmarking -> https://nanonets.com/research/nanonets-ocr-2/
Comment linking /research/nanonets-ocr-2/
on “Open source 3B param model for documents better than Gemini 2.5”Submitted A Practical Guide to OCR with Tesseract as a story · 11 points
HN discussionWe have evaluated against Gemini-2.5-flash. You can check the benchmarks here https://nanonets.com/research/nanonets-ocr-2/#markdown-evalu...
Comment linking /research/nanonets-ocr-2/
on “Show HN: DocStrange: An LLM-Ready Data Platform That Performs Better Than Gemini”Submitted Show HN: Open-source 3B param model better than Mistral OCR as a story · 9 points
HN discussionWe have developed DocStrange to create LLM-ready data from images and PDFs. We have open-sourced a 3B finetuned model also. You can try both the open-sourced and private models from the demo. HF: https://huggingface.co/nanonets/Nanonets-OCR2-3B Demo: https://docstrange.nanonets.com/ Blog: https://nanonets.com/research/nanonets-ocr-2/ This model is an improvement over our last open-source model. We have fixed some of…
Comment linking /research/nanonets-ocr-2/
on “Show HN: DocStrange: An LLM-Ready Data Platform That Performs Better Than Gemini”Submitted Bridging Images and Text – A Survey of VLMs as a story · 9 points
HN discussion
