Feed aggregator

Just launched a document scanner that takes freight documents and combines them with our trust scoring and data to catch fraud.

Comments URL: https://news.ycombinator.com/item?id=48875792

Points: 1

# Comments: 0

Article URL: https://www.washingtonpost.com/technology/2026/07/11/ais-soon-to-be-billionaires-want-know-what-do-with-all-that-money/

Comments URL: https://news.ycombinator.com/item?id=48875773

Points: 1

# Comments: 1

Article URL: https://www.engadget.com/2212928/bluesky-official-ceo-toni-schneider/

Comments URL: https://news.ycombinator.com/item?id=48875772

Points: 1

# Comments: 0

Article URL: https://jeffreylminch.substack.com/p/cuban-missile-crisis-ii-cuban-has

Comments URL: https://news.ycombinator.com/item?id=48875761

Points: 1

# Comments: 1

Article URL: https://github.com/RenderBear/hci

Comments URL: https://news.ycombinator.com/item?id=48875752

Points: 1

# Comments: 1

CozoDB, a transactional relational-graph-vector database with embedded Datalog in Rust, went dormant in December 2024. We hard-forked it as MnesticDB (not official CozoDB), under an MPL-2.0 license, to continue Ziyang Hu and the Cozo Project Authors' vision of building a "Hippocampus for AI", or agentic memory.

But agentic memory isn't a pile of current facts with a log of past decisions. It has to track change over time, and be auditable. We've recently shipped several features that make this possible.

First, we added a distinction between valid time (when a fact is true about the world) and transaction time (when the database came to believe it). This allows time travel, the ability to audit the memory at any point and see what knowledge any past decision was based on. Every write draws its transaction time from a crash-safe monotone commit clock, an atomic high-water mark persisted inside the same transaction that commits the data, so the clock can never advance without the write landing, even across a crash. Because the stamp is allocated in the commit critical section, transaction-time order equals commit order equals visibility order, which makes time travel sound.

Then we implemented the semiring provenance framework from Green et al.'s 2007 paper: the same recursive rule computes existence, cost, confidence, or supporting evidence just by swapping the combine operator, instead of a bespoke tracking system per application. In practice, an aggregate like min_cost_k returns not just an answer but the k best derivations behind it, each with the evidence chain that justifies it. And because those annotations are ordinary values, materializing a derivation into a transaction-time relation composes the two features into an annotated belief history. Every derivation carries the transaction time we came to believe it, so an as-of read tells you not just what we believed at T, but why.

Followed by some more improvements:

- Added ::kill and :timeout to provide the ability to interrupt long-running queries.

- Implemented a deterministic greedy join reorder to prevent naively-ordered conjunctions, the kind that an LLM may write, from spinning. 54.5x improvement with identical results.

- Added an opt-in Yannakakis-style per-key factorized count() instead of join enumeration, with a 4-342x improvement with identical results.

It's available on crates.io and PyPI.

https://crates.io/crates/mnestic https://pypi.org/project/mnestic/

If you're building agent memory and want time-travel + provenance in the graph database rather than bolted on, I'd love to exchange notes.

Special thanks to Matthias Autrata for providing valuable feedback and guidance that was critical to implementing the above improvements.

Comments URL: https://news.ycombinator.com/item?id=48875749

Points: 1

# Comments: 0

Article URL: https://english.elpais.com/usa/2026-07-11/construction-workers-electricians-couriers-how-ice-agents-use-disguises-to-deceive-and-detain-migrants.html

Comments URL: https://news.ycombinator.com/item?id=48875743

Points: 2

# Comments: 2

Article URL: https://ibm-1401.info/pictures/Proc-MIW-2017-Garner-1401PDP1.pdf

Comments URL: https://news.ycombinator.com/item?id=48875738

Points: 1

# Comments: 0

I'm Fenil, co-founder/CEO of OpenFunnel (YC F24), building this with my co-founder/CTO Aditya. We're launching OpenBenchmarks (https://openbenchmarks.com), open-source, reproducible benchmarks for SaaS APIs, starting with the category we know best: GTM APIs.

## Why we built this

More and more B2B software evaluation will/already runs through reasoning models inside agentic workflows rather than through people. And buyers increasingly pick vendors that are API-first and ship MCPs, so they can wire them into internal workflows. Strong reasoning models are skeptical of marketing. When a genuinely neutral benchmark is available, they discount SEO and self-published benchmarks, and they lose trust the moment something looks like a marketing claim.

The thing that survives that skepticism is an independent build-first benchmark the agent can reproduce itself and trust.

Reproducible by default: every cell ships the literal HTTP request/response plus the judge prompt/response

All benchmarks live at https://github.com/openbenchmarks-labs. We start with GTM APIs, with more on the way.

## On Benching ourselves

We benched our own product (OpenFunnel) as a vendor in the lookalikes benchmark, on purpose.

We increasingly saw buyers asking for benchmarks on sales calls and were also curious to see if agents also made decisions the same way.

We wanted to see whether an agent could run the whole loop end to end: discover the benchmark while researching a user's query, weigh it as the deciding factor in picking a winner for that user's category, then sign up and auth through to the chosen vendor to complete the task. That last stretch needs live vendors with agent-auth wired up: us and a few others.

We came out #1 on the current seed (89% vs 74%). Because it's open, we don't win everything: depending on seed and metric (precision@10/@50/@100), we win some and lose others.

## Dogfooding

We dogfooded it and ran 200 incognito simulated buyer flows through Claude Code with live web search, from query to discovery to sign-up, and watched which sources the model fetched and which ones made it into the final decision. The benchmark got opened in a large majority of runs, across the range of queries an actual buyer would ask in Claude Code, from building a lookalikes workflow to picking a lookalike API provider. And when it was opened, it usually drove the final call, over self-published GEO pages and vendor benchmarks with years of domain authority.

We've studied the effect and potential ROI of being benched, we're taking OpenFunnel off the benchmark.

## What's next

More GTM benchmarks, then beyond: devtools and infrastructure.

We're are also working with traditional SaaS companies thinking about going API-first and opening up to a new class of customer: agents

Try it, or hand it to your agent: https://openbenchmarks.com

Comments URL: https://news.ycombinator.com/item?id=48875730

Points: 4

# Comments: 2

Article URL: https://thenewstack.io/entire-git-for-agents/

Comments URL: https://news.ycombinator.com/item?id=48875716

Points: 1

# Comments: 0

Article URL: https://www.thoughtworks.com/en-us/insights/blog/testing/code-review-dead-long-live-code-review

Comments URL: https://news.ycombinator.com/item?id=48875714

Points: 1

# Comments: 0

Article URL: https://github.com/openai/git

Comments URL: https://news.ycombinator.com/item?id=48875709

Points: 21

# Comments: 15

Pretty annoying

Comments URL: https://news.ycombinator.com/item?id=48875680

Points: 1

# Comments: 0

Article URL: https://github.com/mrhustlex/TradingSpy-TradingAgentService

Comments URL: https://news.ycombinator.com/item?id=48875678

Points: 1

# Comments: 0

Article URL: https://xcancel.com/mattshumer_/status/2075657271401390161

Comments URL: https://news.ycombinator.com/item?id=48875670

Points: 3

# Comments: 0

Article URL: https://github.com/sqlsure/sqlsure

Comments URL: https://news.ycombinator.com/item?id=48875342

Points: 1

# Comments: 0

Article URL: https://ai-arcade.app

Comments URL: https://news.ycombinator.com/item?id=48875323

Points: 1

# Comments: 0

Article URL: https://github.com/vvedantb/eva

Comments URL: https://news.ycombinator.com/item?id=48875320

Points: 1

# Comments: 0

Article URL: https://nelson-lang.github.io/nelson-website/

Comments URL: https://news.ycombinator.com/item?id=48875310

Points: 1

# Comments: 0