Alternatives
Products that do what I generated a "stress test" of 200 rare defects from 7 real photos does
Hello HN, I work on vision systems for structural inspection. A common pain point is usually that while we have a lot of "healthy" images, we often lack a reliable "Golden Set" of rare failures (like shattered porcelain) to validate our models before deployment. You can't trust your model's recall if your test set only has 5 examples of the failure mode for example. So to fix this, I built a pipeline to generate datasets. In this example, I took 7 real-world defect samples, extracted their topology/texture, and procedurally generated 200 hard-to-detect variations across different…
- 1IM
Here's a "behind-the-scenes" look at my development of a cool sensor during my PhD (electrical engineering). This sensor is only about 1/3 of my total research for my degree and took about a year. I've been on HN for a while now and I've seen my fair share of posts about the woes of pursuing a PhD. Now that I'm done with mine I wanna share some anecdotal evidence that doing a PhD can actually be enjoyable (not necessarily easy) and also be doable in 3 years. When I started I knew I didn't want to work on something that would never leave the lab or languish in a dissertation PDF no one…
2024 · paulbupejr.com
- 2
- 3LP
Hey! My name is Dima, I am from Ukraine and I am the co-founder of Lost Pixel! We built this tool to solve our problems at work and decided to open-source it so more people can build their custom visual regression testing pipelines! If you want to chat about the tool or visual regression testing in general I am super excited to meet like-minded people! Thanks a lot for checking out the product and I hope it will serve you well if you decide to try it out :D
2022 · github.com
- 4SF
Hi HN, Over the past two years I’ve built and debugged a fair number of production pipelines—mainly retrieval‑augmented generation stacks, agent frameworks, and multi‑step reasoning services. A pattern emerged: most incidents weren’t outright crashes, but silent structural faults that slowly compromised relevance, accuracy, or stability. I began logging every recurring fault in a shared notebook. Colleagues started using the list for post‑mortems, so I turned it into a small public reference: 16 distinct failure modes (semantic drift after chunking, embedding/meaning mismatches,…
2025 · github.com
- 5OS
Hi all! This morning, we released a new Apache 2.0 licensed model on HuggingFace for detecting hallucinations in retrieval augmented generation (RAG) systems. What we've found is that even when given a "simple" instruction like "summarize the following news article," every LLM that's available hallucinates to some extent, making up details that never existed in the source article -- and some of them quite a bit. As a RAG provider and proponents of ethical AI, we want to see LLMs get better at this. We've published an open source model, a blog more thoroughly describing our methodology (and…
2023 · vectara.com
- 6IS
Read an article yesterday about the H.264 codec increasing their licensing fee by an astronomical amount. And as always, my first shot was how hard could it be to try and build a codec which could be that efficient. I've personally been on a drive to improve my ability to one-shot complex features, products, or make even surgical changes. It's been a few months since I've been doing that, and honestly, results have been great for both work and work/life balance. This was a fun experiment. It burned through tokens, but it helped me identify some more improvements I could make to my…
Apr 2026 · github.com
- 77D
hi all. i’ve been shipping a small open project that tries to answer that question with evidence, not vibes. in 70 days it reached \~800 stars. the core claim is simple: many AI failures are not noise. they repeat because the geometry and ordering underneath are stable. if so, we should be able to name each failure mode, set acceptance targets, and stop shipping the same bug twice. ### what it is * a compact Problem Map of 16 reproducible failure modes in RAG and agents. * each item has a minimal fix and measurable gates. examples: * Semantic ≠ Embedding: metric and normalization mismatch.…
2025 · github.com
- 8IB
Hi HN, I've been working on this for a while and it was hard to decide when to stop, either on the way information is presented or when to stop with adding entries. It's not meant as a blog, but rather as a reference that keeps growing. Link: https://thehardparts.dev Currently I've created 4 main section: - Failure Modes: ways project go wrong - Red Flags: early signals that are worth taking seriously - Tech Decisions: common and not so common trade-offs for hard choices - Playbooks - guided approach for situations that repeat I've also focused on creating links between them to…
Apr 2026 · thehardparts.dev
- 9IB
I've gotten to the point in my career where I now make strategic decisions often (hiring, firing, choosing what equipment to go with, etc.), as well as in my personal life where I need to strongly weigh my options for a big purchase or investment. I found a not-so-surprising parallel between the two as these decisions "resolved." Am I making good decisions or am I getting lucky? Did some research, read some books, and realized I should get in the habit of tracking my decision process. That quickly turned into the idea that formed Convexly. The landing page is a 10-question calibration quiz…
Apr 2026 · convexly.app
- 10SR
https://github.com/eigenloss/isthisloss Original thread: https://news.ycombinator.com/item?id=16907615
2018
- 11BI
Hello HN, I built Bulk Image Generator because I was frustrated with how slow it is to iterate on visual assets. Tools like Midjourney are great for single images, but when you need 50+ variations of a product shot or a game asset based on a specific reference, the manual process is a pain. What it does: Upload one reference image and generate 100+ AI variations while maintaining style/structure. Batch background removal (because once you generate 100 images, you usually need to clean them up). Fast bulk download. Why I’m sharing it here: I’m looking for feedback on the consistency of…
Mar 2026 · bulkimagegenerator.app
- 12IB
I had 14,000 photos sitting on a drive and wanted an excuse to play with local vision models and Elixir/Phoenix. I originally tried to get LLaVA to tell me if a photo was 'good' or matched my style, but quickly learned that LLMs have terrible taste. I ended up demoting the LLM to just extract metadata, and built a custom CLIP/Ridge Regression pipeline to actually learn my preferences based on how I rate things. The stack is Phoenix/Oban on the orchestrator side, and Python/FastAPI/Instructor for the AI workers. Happy to answer any questions about the architecture,…
Apr 2026 · qwelian.com
- 13HA
I spent 9 months in abdominal pains. 7 endoscopies. Turned out the problem wasn’t even in my gut. That mess is why I built Healz.ai - AI + real doctors-investigator who build complete picture and look for root causes. I’m not here to pitch. I just want to know if it’s real or crap and are there people like me sometime ago. Please burn me as much as possible. Just want to make no bullshit service. No marketing. No “AI doctor”, NO "text-to-porn generator" hype. Just trying to make something that could save people the months I lost.
Oct 2025 · app.healz.ai
- 14IW
I've spent years watching hardware teams make the same mistakes — tooling committed before thermal was resolved, manufacturing processes not validated in PVT that only showed defects at volume, components going EOL unnoticed until the factory tried to order them. The pattern was always the same. The costs were always avoidable. Nobody had written it down. The result is The Hardest Hardware Lessons — covering the full journey from first prototype to end-of-life: EVT/DVT/PVT, factory evaluation, supply chain risk, mass production operations, field failures, and product lifecycle…
Mar 2026 · thehardesthardwarelessons.com
- 15NB
I've spent weeks curating technical implementation details of how companies are actually deploying LLMs and Generative AI in production. The database now contains over 300 case studies with detailed technical summaries (230,000+ words) focusing exclusively on architectural decisions, deployment patterns, and real engineering challenges. Key features: * Each case study is technically focused - no marketing fluff * 150+ entries from technical conference talks and panels (saving you 100+ hours of video watching) * Sophisticated filtering by technical stack, RAG implementations, monitoring…
2024 · zenml.io
- 16QT
Hey HN, Screenshots of bugs are no substitute for reproducing issues, but clearly annotated/laid out images do go a long way for developers to understand what’s going on and start working on them sooner. These remain an odd amount of work/overhead to do well— no one is going to fire up photoshop here, and most QA folks I’ve talked to have a simple trick or two up their sleeve for assembling screenshots/repro content quickly. A friend of mine’s spent the last few years working to improve this process and threw together what’s quite possibly the most functional screenshot tool…
2022
- 17SI
Hey guys, I would really appreciate it if you could take a look at my new site that I have put into private beta this weekend. In summary it is a site for photographers, graphic designers and web designers to get private feedback on their work. There are plenty of sites out there were we can post our work and get it seen but sometimes you need to be able to get some feedback before having it on a public display. This is part 1 of the longer term plan. The beta has the basics to enable users to review each others work, the number of images are limited at the minute but I will be extending…
2011
- 18G4
When new Large Multimodal Models (LMMs) are released, there is excitement as we explore new capabilities. What can a model do? What can't a model do? What strange behaviors does the model exhibit? With that said, such analyses are frozen in time. At a hackathon toward the end of last year, the Roboflow team made a tool that runs the same set of tests with the GPT-4 with Vision API every day. This allows people to see how the model performs over time as updates are made. The last seven days of results are displayed on a web page; the rest of the data is archived in GitHub. We started the site…
2024 · gptcheckup.com
- 19CC
Hey HN, I've been wanting to start this project for a long time, ever since I read an article about collaborative transformation of blob images[1], and I'm hoping it will take off here as I'm curious to see how it can grow. Here's how it works: - My website is available at https://chaospixels.com. You can find its code in the `src` directory of the linked repository. - Every day, I'll review all the open PRs (Pull Requests) and calculate a PR score based on the number of thumbs-up on the PR's opening comment. - The PR with the highest score will be merged and deployed. - Afterward,…
2024 · github.com
- 20TN
Hi guys, I’m excited to share an update on ReproModel, an open-source toolbox designed to streamline the testing and reproduction of machine learning models. I, like many of you, have really struggled with benchmarking and comparing models, from missing code, to opaque experiment parameters slowing the process. I decided to take matters into my own hands, and created a mini-toolbox in my free time to streamline the process. The goal is to reduce the time and effort spent on replicating experiments, enabling researchers to focus on innovation rather than setup. Knowing this task is not an…
2024 · github.com
- 21BA
I'm 17 and recently launched a private beta for Biohack - a tool that scans food products and gives a longevity score based on impact on aging factors such as inflammation, omega ratio, toxins and more. Still early and a bit rough, but I'm looking for honest feedback from people who care about health span and deep nutrition. Would love your thoughts.
2025 · getbiohack.app
- 22
- 23IB
Hello everyone, I doubt this would be relevant to the kind of person who uses HN, but I thought I could share for some feedback. I built this site because there is a whole world of people who believe in new age spirituality and I am very much one of them. It is a site where you get the users gender their goals and their images and use AI and psychology to generate images of them in the process of achieving their goals. I am so deeply struggling with how to get this highly on Google. I don't even know if that is important anymore. What are your suggestions with distribution and getting in…
2025 · visionboardsai.com
- 24IE
Hey HN, when building ML systems for industrial AI, we have learned that data inspection is critical during the ML development process. We are also big fans of the Hugging Face ecosystem. That is why we built an integration to our data exploration tool Spotlight that allows you to interactively explore Hugging Face datasets with one line of code. Spotlight lets you leverage model results such as predictions and embeddings to gain a deeper understanding in data segments and model failure modes. Currently, many many NLP, CV, Audio and multimodal datasets are supported both locally and on the…
2023 · huggingface.co
Ranked by how close each launch is in meaning, then by votes. Refine with a description →