Dev Systems
Show HN: LMM for LLMs – A mental model for building LLM apps
I've been building agentic apps for some large Fortune 500 companies (T-Mobile, Twilio, etc.) and developed a mental model that serves as a practical guide in building agentic apps: separate the high-level agent specific logic from low-level platform capabilities. I call it the L-MM: the Logical Mental Model for LLM applications.This mental model has not only been tremendously helpful in building agents but also helping customers think about the development process - so when I am done with
Show HN:[Opensource] AIgr.id–Polycentric Infrastructure for Open and Plural AI
Hey HN! I'm Kanishka Nithin, founder of AIGr.id (https://www.aigr.id).
We’re building AIGr.id — a polycentric network of independent, modular AI that can coordinate, exchange data, and compose into higher-level intelligence — all within a decentralized and plural ecosystem. Rings collective intelligence?In simpler terms: We’re trying to make it possible for people to produce, remix, operate, distribute and consume AI systems the way we use the internet— openly, collaboratively, a
Show HN: Developer First Observability with Prism
Over the past few months, we've been building Prism; a fast, developer-first observability UI that's deeply integrated with our open-source database, Parseable. Prism is designed for engineers who debug distributed systems, sometimes at odd hours, and need answers right now, without having to deal with five different tools.Backstory:We spoke to dozens of observability practitioners in large enterprises. Across the board, we heard two things:(1)The RoI on observability tools is hard to
Show HN: Aigr.id – Decentralized Internet of Intelligence
Hey HN,I'm Narasimha Prasanna, co-founder of AIGr.id. We're excited to introduce AIGr.id, a decentralized, community-driven global AI network designed as public infrastructure—think of it as the Internet of Intelligence.Instead of large, closed, monolithic AI systems, AIGr.id interconnects smaller, independent AI clusters through AIOS, a decentralized AI operating system that's 100% open-source.Today’s AI landscape is:- Centralized: Resource-heavy systems demand vast funding, comp
Show HN: A new way to verify remote AI model execution (no TEEs, no ZK)
We’ve published a whitepaper proposing a simple method to verify that a remote AI node is executing a specific model — without using trusted execution environments (TEEs) or zero-knowledge proofs.The method relies on a minimal set of reference outputs to test model behavior, enabling trust without full re-execution.Curious to hear thoughts from folks working in ML, distributed systems, or trustless compute.https://arxiv.org/abs/2504.13443
Ask HN: Programming career prospects for beginners in regards to AI
I'm a beginner who enjoys programming, particularly the larger scale side of software and systems architecture. I'm trying to understand how AI might impact programming careers in the next 5-10 years.Some specific questions I'm curious about:From those working in the industry, how has generative AI (I've heard great things about Gemini 2.5 but haven't tested it out yet) currently affected your work and the general software engineering landscape?Do you think programming w
How we built Cloud WAN to support businesses amid the rise of AI
<img src="https://storage.googleapis.com/gweb-uniblog-publish-prod/images/CloudWAN_Hero_2097x1182.max-600x600.format-webp.webp">Go behind the scenes on the development of Google Cloud WAN, now available to external customers for the first time.
Untangling fan engagement: Carlo De Marchis (A Guy with a Scarf) on going beyond the raw metrics
Fan engagement strategies are evolving rapidly. But how do organizations ensure they are delivering realtime data with real meaning to their customers? As a veteran of the sports media industry with 35 years of experience at Deltatre, Carlo De Marchis (A Guy With a Scarf) recently gave a keynote at Ably’s fan engagement summit in February, and explored exactly this question - what should companies actually be trying to achieve when building a fan engagement strategy? His answer was concise: In
Authentication and authorization in a microservice architecture: Part 1 - Introduction
This article is the first in a series of articles about authentication and authorization in a microservice architecture.The complete series is: Overview of authentication and authorization in a microservice architecture Implementing authentication Implementing simple authorization Developing complex authorization Implementing complex authorization using Oso Cloud - part 1 Implementing complex authorization using Oso Cloud - part 2 Authentication is the me
The latest Workspace feature drop includes Gmail updates, Gemini tools and more.
<img src="https://storage.googleapis.com/gweb-uniblog-publish-prod/images/April_drops_thumbnail_1.max-600x600.format-webp.webp">Our latest Google Workspace feature drop is here with new tools to help you collaborate, stay on top of your to-do list and do your best work. Plus, this latest update h…
Show HN: I made an open source Idea to App WebApp
Create MVPs with AI in seconds.A comprehensive guide to building Minimum Viable Products using the latest AI tools and technologies. Generate complete implementation plans in minutes.Comes with multiple add-ons:
FREE Tools for EVERYONE: Yes, completely free for all users!AI Chat: Brainstorm, refine ideas, and get instant feedback by chatting directly with powerful AI models like Gemini 1.5 Flash & OpenAI (more coming!). Preserve context across sessions!AI Tool Comparison: Stop hunting throu
Empower your teams with modern architecture governance
Agile product teams thrive on autonomy and rapid iteration, especially in the cloud where they can quickly deploy and test systems. However, traditional architecture governance often stands in their way, because many enterprises still impose centralized, one-off architecture signoffs early in the process. Historically, these signoffs verified design compliance with corporate standards in a slower, on-premises world. In cloud environments, such signoffs quickly become obsolete—along with their as
Distributed P2P LinkedIn query system
Linkedin (free version) is extremely search restrictive for a social network, economically makes sense for them with all the professional data/ networks, but i'm tired of making inroads for finding data on what career paths are available only to be popped up with irritating "limited search results to be viewed" or "opt for Linkedin premium" to view more profile, etc.As a result was wondering if we could build a distributed query system (p2p) which leverages user
Show HN: Goldbach Conjecture up to 4*10^18+7*10^13
Achieved a new world record in verifying the Goldbach Conjecture using grid computing, by extending the verification up to 4 quadrillion (4×10¹⁸) + 70 trillion (7×10¹³).My grid computing system - Gridbach is a cloud-based distributed computing system accessible from any PC or smartphone. It requires no login or app installation. The high-performance WASM (WebAssembly) binary code is downloaded as browser content, enabling computation on the user’s browser.[Website]
https://gridbach.com
Every web application that may need scale later should be a distributed system
If you know your web application will not need scale ever, then you don't need a distributed system.However, learning distribtued system design is not that complicated unless you are an entry-level developer.I guess people who tell others to not care about scalability for a software startup are probably marketers, hustlers, or financiers who consider product building as someone else's responsibility and are willing to discard failing products and hire developers for a new product. They
I built a Semantic Operating System: KounGraph – a second brain made of meaning
Over the past year, I've been designing a system that tries to rethink how we structure thought, memory, tasks, AI interaction, and human–AI collaboration. The result is something I call KounGraph, a semantic operating system based on the idea that everything is a node—not just notes, but relationships, reasoning steps, decisions, tasks, even the AI that talks to you.Instead of text and tags, it uses semantic nodes with source metadata, relationships-as-nodes, recursive logic, and AI reason
Show HN: Unbody is open-source – The Supabase of the AI era
We're open-sourcing Unbody, a backend framework that simplifies building AI-native applications. Instead of having to manually stitch together vector databases, embedding models, and LLM wrappers, Unbody provides a unified system with sensible defaults that you can customize.The Problem:
Building AI applications today means making dozens of infrastructure decisions:- Which vector store to use?
- How to handle data ingestion and chunking?
- Which embedding model to choose?
- How to structure
Ask HN: Is 'vibe coding' the future of software development?"
I've been thinking about the effects of 'vibe coding' on software development.Imagine a two-phase approach: First, business users work with AI coding agents to specify and create functional prototypes without writing code ("vibe coding"), then developers take over to rebuild and productionize these prototypes.This could potentially:- Speed up initial development by bypassing traditional requirements gathering- Allow business users to directly express their needs as worki
Ask HN: Year Deep Dive in ML
I'm planning to spend significant time this year on ML and I'm looking for advice, feedback from people that did it successfully.I have quite a lot (>10Y) of experience in software and everything data already, and at some distant point in the past I did play with NNs a bit - I have a decent (for a non practitioner) understanding - but that was before LLM, transformers.I'm not necessarily looking to learn things that could get me a job in the field in the short term, but much mo
Show HN: Strain Seer – From demo-day hack to engineering blueprint
Hi HN! I'm Keming.Hackathons have a throwaway-project culture. When my strain analysis (research) tool won 3rd place at Make OHI/O, I decided to break that pattern.During the actual 24-hour competition, I pushed beyond the usual expectation. I built a working app with full test coverage and deployed it live. But after everyone packed up and went home? That's when the real work began.I challenged myself to transform this quick hack into production-grade software. I built CI/CD