Show HN: US Publicly Traded Companies probabilities of default with public data https://ift.tt/aolQMih

Show HN: US Publicly Traded Companies probabilities of default with public data https://ift.tt/v01Kg9m November 14, 2025 at 03:51AM

Show HN: YAML Validator –A simple Docker-based YAML checker https://ift.tt/27y0Uqm

Show HN: DBOS Java – Postgres-Backed Durable Workflows https://ift.tt/pfINcHu

Show HN: DBOS Java – Postgres-Backed Durable Workflows Hi HN - I’m Peter, here with Harry (devhawk), and we’re building DBOS Java, an open-source Java library for durable workflows, backed by Postgres. https://ift.tt/TwuEYZq Essentially, DBOS helps you write long-lived, reliable code that can survive failures, restarts, and crashes without losing state or duplicating work. As your workflows run, it checkpoints each step they take in a Postgres database. When a process stops (fails, restarts, or crashes), your program can recover from those checkpoints to restore its exact state and continue from where it left off, as if nothing happened. In practice, this makes it easier to build reliable systems for use cases like AI agents, payments, data synchronization, or anything that takes hours, days, or weeks to complete. Rather than bolting on ad-hoc retry logic and database checkpoints, durable workflows give you one consistent model for ensuring your programs can recover from any failure from exactly where they left off. This library contains all you need to add durable workflows to your program: there's no separate service or orchestrator or any external dependencies except Postgres. Because it's just a library, you can incrementally add it to your projects, and it works out of the box with frameworks like Spring. And because it's built on Postgres, it natively supports all the tooling you're familiar with (backups, GUIs, CLI tools) and works with any Postgres provider. If you want to try it out, check out the quickstart: https://ift.tt/g459GAI We'd love to hear what you think! We’ll be in the comments for the rest of the day to answer any questions. https://ift.tt/TwuEYZq November 14, 2025 at 02:03AM

Show HN: LLM fine-tuning without infra or ML expertise (early access) https://ift.tt/46f0cCq

Show HN: LLM fine-tuning without infra or ML expertise (early access) https://www.tinytune.xyz/ November 14, 2025 at 12:33AM

We Want to Hear Your Transportation Priorities: New Survey and Budget Planning Update

We Want to Hear Your Transportation Priorities: New Survey and Budget Planning Update
By Caroline Cabral

Our budget planning process is underway, and Muni is approaching the largest budget crisis transit has ever faced. We’re not alone – transit systems across the country are grappling with impending fiscal cliffs. As we work to make our services and spending more efficient, and identify funding to bridge Muni’s budget gap, we want to hear from local communities about their transportation priorities. This will help us understand which services are most essential to maintain. Some of the transportation needs the SFMTA supports include: Fast, frequent and reliable Muni service Repairs and...

Published November 12, 2025 at 05:30AM
https://ift.tt/YH8QO0k

Show HN: JavaScript Engines Zoo https://ift.tt/v4DnVBu

Show HN: JavaScript Engines Zoo https://ift.tt/RYd4Vr2 November 12, 2025 at 09:32PM

Show HN: Cancer diagnosis makes for an interesting RL environment for LLMs https://ift.tt/legimbw

Show HN: Cancer diagnosis makes for an interesting RL environment for LLMs Hey HN, this is David from Aluna (YC S24). We work with diagnostic labs to build datasets and evals for oncology tasks. I wanted to share a simple RL environment I built that gave frontier LLMs a set of tools that lets it zoom and pan across a digitized pathology slide to find the relevant regions to make a diagnosis. Here are some videos of the LLM performing diagnosis on a few slides: ( https://www.youtube.com/watch?v=k7ixTWswT5c ): traces of an LLM choosing different regions to view before making a diagnosis on a case of small-cell carcinoma of the lung ( https://youtube.com/watch?v=0cMbqLnKkGU ): traces of an LLM choosing different regions to view before making a diagnosis on a case of benign fibroadenoma of the breast Why I built this: Pathology slides are the backbone of modern cancer diagnosis. Tissue from a biopsy is sliced, stained, and mounted on glass for a pathologist to examine abnormalities. Today, many of these slides are digitized into whole-slide images (WSIs)in TIF or SVS format and are several gigabytes in size. While there exists several pathology-focused AI models, I was curious to test whether frontier LLMs can perform well on pathology-based tasks. The main challenge is that WSIs are too large to fit into an LLM’s context window. The standard workaround, splitting them into thousands of smaller tiles, is inefficient for large frontier LLMs. Inspired by how pathologists zoom and pan under a microscope, I built a set of tools that let LLMs control magnification and coordinates, viewing small regions at a time and deciding where to look next. This ended up resulting in some interesting behaviors, and actually seemed to yield pretty good results with prompt engineering: - GPT 5: explored up to ~30 regions before deciding (concurred with an expert pathologist on 4 out of 6 cancer subtyping tasks and 3 out of 5 IHC scoring tasks) - Claude 4.5: Typically used 10–15 views but similar accuracy as GPT-5 (concurred with the pathologist on 3 out of 6 cancer subtyping tasks and 4 out of 5 IHC scoring tasks) - Smaller models (GPT 4o, Claude 3.5 Haiku): examined ~8 frames and were less accurate overall (1 out of 6 cancer subtytping tasks and 1 out of 5 IHC scoring tasks) Obviously, this was a small sample set, so we are working on creating a larger benchmark suite with more cases and types of tasks, but I thought this was cool that it even worked so I wanted to share with HN! November 12, 2025 at 10:31PM