Outcome-Based Reinforcement Learning to Predict the Future
13 by bturtel | 0 comments on Hacker News.
13 by bturtel | 0 comments on Hacker News.
Why Cline Doesn't Index Your Codebase (and Why That's a Good Thing)
45 by intrepidsoldier | 29 comments on Hacker News.
45 by intrepidsoldier | 29 comments on Hacker News.
Show HN: Malai – securely share local TCP services (database/SSH) with others
12 by amitu | 7 comments on Hacker News.
malai is a peer to peer network, and is a dead simple to share your local development HTTP server, without setting up tunnels, dealing with firewalls, or relying on cloud services. In malai 0.2.5, we have added TCP support, which means you can expose any TCP service to others using malai, without opening the TCP service related port to Internet. With malai installed on both ends, any TCP service can be securely tunneled over it. It can be used to secure your SSH service, or securely share your database server. GitHub: https://ift.tt/tJ684SP (star us!) Would love feedback, questions, or ideas — thanks! PS: We have also added `malai folder`, which lets you share (readonly) the content of a folder with others.
12 by amitu | 7 comments on Hacker News.
malai is a peer to peer network, and is a dead simple to share your local development HTTP server, without setting up tunnels, dealing with firewalls, or relying on cloud services. In malai 0.2.5, we have added TCP support, which means you can expose any TCP service to others using malai, without opening the TCP service related port to Internet. With malai installed on both ends, any TCP service can be securely tunneled over it. It can be used to secure your SSH service, or securely share your database server. GitHub: https://ift.tt/tJ684SP (star us!) Would love feedback, questions, or ideas — thanks! PS: We have also added `malai folder`, which lets you share (readonly) the content of a folder with others.
Pyrefly vs. Ty: Comparing Python's Two New Rust-Based Type Checkers
17 by edwardjxli | 2 comments on Hacker News.
17 by edwardjxli | 2 comments on Hacker News.
Launch HN: Relace (YC W23) – Models for fast and reliable codegen
9 by eborgnia | 3 comments on Hacker News.
Hey HN community! We're Preston and Eitan, and we're building Relace ( https://relace.ai ). We're trying to make building code agents easy and cheap. Here’s an example of our apply model vs. whole file edits: https://youtu.be/J0-oYyozUZw Building reliable code agents is hard. Beyond simple prototypes, any app with code generation in production quickly runs into two problems -- how do you reliably apply diffs, and how do you manage codebase context? We're focused on solving these two problems at order-of-magnitude lower price and latency. Our first model that we released, in February, is the Fast Apply model -- it merges code snippets with files at 4300 tok/s. It is more reliable (in terms of merge errors) than Sonnet, Qwen, Llama, or any other model at this task. Each file takes ~900ms and gives an instantaneous user experience, as well as saving ~40% on Claude 4 output tokens. Our second model focuses on retrieval. For both vibe-coded and enterprise codebases, retrieving only the files relevant to a user request saves both on SoTA input token cost and reduces the number of times code agents need to view files. Our reranker (evals below) can scan a million-line codebase in ~1-2s, and our embedding model outperforms any other embedding model for retrieval as evaluated on a corpus of Typescript/React repositories. There are many different ways to build coding agents, but being able to edit code reliably and retrieve the most relevant parts of the codebase is going to be a foundational issue. We're excited to be building ways to make it more accessible to millions of users who don't want to spend $$$ on Claude. These models are used in production, millions of times per week. If you've used Lovable, Create.xyz, Magic Patterns, Codebuff, Tempo Labs then you've used us! Here's a link to try it out: https://app.relace.ai , and here are our docs: https://docs.relace.ai . We've opened up free access for prototyping on our website to everyone, and the limits should be enough for personal coding use and building small projects (correct us if it’s not). We integrate directly with Open-Source IDE's like Continue.dev. Please try us out, we'd love to hear your feedback!
9 by eborgnia | 3 comments on Hacker News.
Hey HN community! We're Preston and Eitan, and we're building Relace ( https://relace.ai ). We're trying to make building code agents easy and cheap. Here’s an example of our apply model vs. whole file edits: https://youtu.be/J0-oYyozUZw Building reliable code agents is hard. Beyond simple prototypes, any app with code generation in production quickly runs into two problems -- how do you reliably apply diffs, and how do you manage codebase context? We're focused on solving these two problems at order-of-magnitude lower price and latency. Our first model that we released, in February, is the Fast Apply model -- it merges code snippets with files at 4300 tok/s. It is more reliable (in terms of merge errors) than Sonnet, Qwen, Llama, or any other model at this task. Each file takes ~900ms and gives an instantaneous user experience, as well as saving ~40% on Claude 4 output tokens. Our second model focuses on retrieval. For both vibe-coded and enterprise codebases, retrieving only the files relevant to a user request saves both on SoTA input token cost and reduces the number of times code agents need to view files. Our reranker (evals below) can scan a million-line codebase in ~1-2s, and our embedding model outperforms any other embedding model for retrieval as evaluated on a corpus of Typescript/React repositories. There are many different ways to build coding agents, but being able to edit code reliably and retrieve the most relevant parts of the codebase is going to be a foundational issue. We're excited to be building ways to make it more accessible to millions of users who don't want to spend $$$ on Claude. These models are used in production, millions of times per week. If you've used Lovable, Create.xyz, Magic Patterns, Codebuff, Tempo Labs then you've used us! Here's a link to try it out: https://app.relace.ai , and here are our docs: https://docs.relace.ai . We've opened up free access for prototyping on our website to everyone, and the limits should be enough for personal coding use and building small projects (correct us if it’s not). We integrate directly with Open-Source IDE's like Continue.dev. Please try us out, we'd love to hear your feedback!
I Salvaged $6k of Luxury Items Discarded by Duke Students
29 by drvladb | 20 comments on Hacker News.
29 by drvladb | 20 comments on Hacker News.
US pauses new student visa interviews as it mulls expanding social media vetting
39 by spenvo | 15 comments on Hacker News.
39 by spenvo | 15 comments on Hacker News.
Show HN: Free mammogram analysis tool combining deep learning and vision LLM
8 by coolwulf | 7 comments on Hacker News.
I've built Neuralrad Mammo AI, a free research tool that combines deep learning object detection with vision language models to analyze mammograms. The goal is to provide researchers and medical professionals with a secondary analysis tool for investigation purposes. Important Disclaimers:- NOT FDA 510(k) cleared - this is purely for research investigation- Not for clinical diagnosis - results should only be used as a secondary opinion- Completely free - no registration, no payment, no data retention What it does:1. Upload a mammogram image (JPEG/PNG)2. AI identifies potential masses and calcifications3. Vision LLM provides radiologist-style analysis4. Interactive viewer with zoom/pan capabilities You can try it with any mass / calcification mammo images, e.g. by searching Google: mammogram images mass Key Features:- Detects and classifies masses (benign/malignant)- Identifies calcifications (benign/malignant) - Provides confidence scores and size assessments- Generates detailed analysis using vision LLM- No data storage - images processed and discarded Use Cases:- Medical research and education- Second opinion for researchers- Algorithm comparison studies- Teaching tool for radiology training- Academic research validation The system is designed specifically for research investigation purposes and to complement (never replace) professional medical judgment. I'm hoping this can be useful for the medical AI research community and welcome feedback on the approach. Address: https://ift.tt/icWqFB2
8 by coolwulf | 7 comments on Hacker News.
I've built Neuralrad Mammo AI, a free research tool that combines deep learning object detection with vision language models to analyze mammograms. The goal is to provide researchers and medical professionals with a secondary analysis tool for investigation purposes. Important Disclaimers:- NOT FDA 510(k) cleared - this is purely for research investigation- Not for clinical diagnosis - results should only be used as a secondary opinion- Completely free - no registration, no payment, no data retention What it does:1. Upload a mammogram image (JPEG/PNG)2. AI identifies potential masses and calcifications3. Vision LLM provides radiologist-style analysis4. Interactive viewer with zoom/pan capabilities You can try it with any mass / calcification mammo images, e.g. by searching Google: mammogram images mass Key Features:- Detects and classifies masses (benign/malignant)- Identifies calcifications (benign/malignant) - Provides confidence scores and size assessments- Generates detailed analysis using vision LLM- No data storage - images processed and discarded Use Cases:- Medical research and education- Second opinion for researchers- Algorithm comparison studies- Teaching tool for radiology training- Academic research validation The system is designed specifically for research investigation purposes and to complement (never replace) professional medical judgment. I'm hoping this can be useful for the medical AI research community and welcome feedback on the approach. Address: https://ift.tt/icWqFB2
Running GPT-2 in WebGL: Rediscovering the Lost Art of GPU Shader Programming
19 by nathan-barry | 5 comments on Hacker News.
19 by nathan-barry | 5 comments on Hacker News.
Show HN: Maestro – A Framework to Orchestrate and Ground Competing AI Models
6 by defqon1 | 0 comments on Hacker News.
ive spent the past few months designing a framework for orchestrating multiple large language models in parallel — not to choose the “best,” but to let them argue, mix their outputs, and preserve dissent structurally. It’s called Maestro heres the whitepaper https://ift.tt/XuFsdlH (Narrative version here: https://ift.tt/eC9LBx6... ) Core ideas: Prompts are dispatched to multiple LLMs (e.g., GPT-4, Claude, open-source models) The system compares their outputs and synthesizes them It never resolves into a single voice — it ends with a 66% rule: 2 votes for a primary output, 1 dissent preserved Human critics and analog verifiers can be triggered for physical-world confirmation (when claims demand grounding) The feedback loop learns not only from right/wrong outputs, but from what kind of disagreements lead to deeper truth Maestro isn’t a product or API — it’s a proposal for an open, civic layer of synthetic intelligence. It’s designed for epistemic integrity and resistance to centralized control. Would love thoughts, critiques, or collaborators.
6 by defqon1 | 0 comments on Hacker News.
ive spent the past few months designing a framework for orchestrating multiple large language models in parallel — not to choose the “best,” but to let them argue, mix their outputs, and preserve dissent structurally. It’s called Maestro heres the whitepaper https://ift.tt/XuFsdlH (Narrative version here: https://ift.tt/eC9LBx6... ) Core ideas: Prompts are dispatched to multiple LLMs (e.g., GPT-4, Claude, open-source models) The system compares their outputs and synthesizes them It never resolves into a single voice — it ends with a 66% rule: 2 votes for a primary output, 1 dissent preserved Human critics and analog verifiers can be triggered for physical-world confirmation (when claims demand grounding) The feedback loop learns not only from right/wrong outputs, but from what kind of disagreements lead to deeper truth Maestro isn’t a product or API — it’s a proposal for an open, civic layer of synthetic intelligence. It’s designed for epistemic integrity and resistance to centralized control. Would love thoughts, critiques, or collaborators.
Why the Original Macintosh Had a Screen Resolution of 512×324
5 by ingve | 0 comments on Hacker News.
5 by ingve | 0 comments on Hacker News.
In Vietnam, an unlikely outpost for Chicano culture
5 by donnachangstein | 1 comments on Hacker News.
5 by donnachangstein | 1 comments on Hacker News.
Show HN: My LLM CLI tool can run tools now, from Python code or plugins
16 by simonw | 3 comments on Hacker News.
16 by simonw | 3 comments on Hacker News.