At work we've been prompted about running Claude Code overnight. The suggestion came in form of a document that loosely outlined how this could be done...
#autonomous-coding
65 items
Overnight autonomous coding (www.reddit.com) How to Build an AI Software Factory: Agents That Open, Review, and Merge PRs (www.firecrawl.dev via hn) An AI software factory runs fleets of autonomous coding agents that open, review, and merge PRs. How Stripe, Spotify, Shopify, Uber, and Ramp built theirs, broken into five stages you can implement.
Show HN: Singular – local orchestration for autonomous coding agents (github.com via hn) https://github.com/user-attachments/assets/826bf112-645b-4293-8b63-cf1c8ad6038b singular Autonomous multi-agent orchestration for software repos. One engine, many consumers.
Does Code Cleanliness Affect Coding Agents? (arxiv.org via hn) As autonomous coding agents see rapid adoption, their evaluation has primarily focused on task completion rates holding the target codebase fixed. This leaves a critical question unanswered: does the structural and stylistic quality, or ``…
Testing Local LLMs in Practice: Code Generation, Quality vs. Speed (www.reddit.com) Hello, I spent the last few months building an AI agent that autonomously writes Go code using local LLMs. The primary use case is log parser generation for SIEM pipelines.
Ask HN: Are tab completion models dead? (news.ycombinator.com) I think there is still place for this type of coding where you autocomplete predictable chunks and review it as you go. But I haven't seen much development in this space.
GPT-6 Astra is generally available in GitHub Copilot (github.blog via hn) GPT-6 Astra is generally available in GitHub Copilot GPT-6 Astra from OpenAI is now available in GitHub Copilot. OpenAI’s latest general-purpose model, GPT-6 Astra, is designed for long-horizon, autonomous coding and agentic tasks.
↯ Copilot↯ GPT-6 Astra Releaseautonomous-codingcopilotopenai+1
Miii – Claude Code-level terminal workflows offline, no API keys (www.npmjs.com via hn) The local-first autonomous coding agent. Claude Code UX powered by Ollama.
Coding agents ranked for builders shipping AI software (rankings.leffke.com via hn) AI coding agents / IDE copilots IDE copilots and autonomous coding agents that edit, plan, and ship code with you. ← All categories Organic ranking Free: overall score + blurb.
To What Extent Does Agent-Generated Code Require Maintenance? An Empirical Study (arxiv.org via hn) LLM-based autonomous coding agents have reshaped software development. While these agents excel at code generation, open questions persist about the long-term maintainability of AI-generated code.
Ask HN: How do you gate an autonomous coding agent's shell access? (news.ycombinator.com) I've been giving coding agents more autonomy lately, letting them run shell commands unattended for longer stretches, and I don't have a good answer for how people actually gate that beyond "run it in a container and hope." A container lim…
AVO: Agentic Variation Operators for Autonomous Evolutionary Search (arxiv.org via hn) Agentic Variation Operators (AVO) are a new family of evolutionary variation operators that replace the fixed mutation, crossover, and hand-designed heuristics of classical evolutionary search with autonomous coding agents. Rather than con…
Show HN: Artifex - Graph Based GPU Harness for AI Agents (gatewai.studio via hn) Artifex is a machine-first, headless CLI runtime built for autonomous coding agents to author, validate, and render media node graphs locally. Similar to Deepseek harness released yesterday; which ground models in modular execution environ…
Coding agents ignore open source contribution guidelines, researchers find (thenewstack.io via hn) Coding agents ignore open source contribution guidelines, researchers find. Autonomous coding agents ignore contribution rules in open source communities, finds a new study from researchers at Peking University.
A 15-day autonomous coding run spent five days building no product code (github.com via hn) I told Codex to prove everything. And the proof ate the project.
Trust but Verify? Uncovering the Security Debt of Autonomous Coding Agents (arxiv.org via hn) The increasing adoption of autonomous coding agents accelerates software development but also introduces scoped security risks within high-impact file paths that can outpace traditional human review capacity. While prior research has prima…
Beekeeper, threat intelligence for autonomous coding agents (beekeeper.vercel.app via hn) Web site created using create-react-app
"China Outpacing Us on AI": Why Musk Bought an Autonomous Coding Startup (en.sedaily.com via hn) Elon Musk's SpaceX acquired Cursor parent Anysphere for $60 billion, betting on developer-focused AI amid rising token costs and surging Chinese open-source model use.
Aegis OS: a transactional multi-tenant OS runtime for autonomous coding agents (arxiv.org via hn) Autonomous coding agents such as Google Antigravity (agy), Anthropic Claude Code (claude), and OpenAI Codex CLI (codex) execute multi-step software engineering workflows on workstations and cloud hosts. Existing agent environments face a S…
Show HN: I built a safety gate for autonomous coding agents (inbin-gate.vercel.app via hn) Safe auto-approve for IBM Bob: an agent may act on its own only on values someone with authority actually stated. Live decision demo running the published resolver in your browser.
IDHN – Task-scoped filesystem permissions for AI coding agents (github.com via hn) IDHN v0.3.1 -- Interchangeable Deterministic Host Network IDHN enforces task-scoped permissions for autonomous coding agents. In a controlled experiment, it prevented an unauthorized modification inside the project that the tested native w…
A course on adapting agile when agents write most of the code (learnvoro.com via hn) Agent-Driven Agile: Engineering Process Design for Small Startup Teams Learn to architect and run a streamlined agile development process tailored for a seven-engineer startup team leveraging autonomous coding agents. By the end of this co…
Ask HN: How do you cap what your coding agents can spend? (news.ycombinator.com) I run autonomous coding agents and I'm trying to understand how other people handle money when an agent needs to pay for things (API credits, cloud, domains, SaaS). I've read the earlier "agents that spend real money" thread and the storie…
Agentic coding is a financial trap (frugaast.dev via hn) Agentic coding is a financial trap The hidden financial unit economics of pay-per-token autonomous coding loops. The era of predictable $20-a-month AI subscriptions is quietly ending, replaced by autonomous tools that silently burn through…
Practical Secrets Extraction Against Black-Box LLMs (arxiv.org via hn) Large language models (LLMs) increasingly power autonomous coding agents such as Codex and Claude Code, yet their training corpora may contain confidential credentials exposed in public repositories or collected from private development ar…
An AI coding agent silently erased 92% of AI nodes in n8n's most-cited dataset (sevenedge.pl via hn) An autonomous coding agent's commit quietly erased the real configuration of 92.6% of AI agent nodes in n8n's most-cited public workflow dataset.
Hours unattended: the memory bugs that broke my autonomous coding agent (eltoncherrington.github.io via hn) A field report on running a coding agent unattended for three days: a silently-dropped password, a truncation cascade, a clock that posted into the future, and the wrong-proofing that fixed them.
Real world(ish) DeepSeek V4 Flash performance on a single MI300X (matthusby.github.io via hn) DeepSeek V4 Flash performance on a single MI300X What a single GPU actually delivers when the clients are autonomous coding agents doing real work, not synthetic benchmark load. Built on the open-source deepseek-v4-flash-mi300x serving sta…
Show HN: Claw-coder- the first autonomous local first coding agent (openmindedai.org via hn) Hello everyone , my name is Gabriel Blessed and I have been working in claw-coder for 6 months but I have wanted real user feedback on claw-coder, claw-coder is a local autonomous coding agent that lives in your terminal and I have built i…
Modelio 6.2 ported to native Apple Silicon ARM64 with Codex (github.com via hn) This work adds a native Apple Silicon port of Modelio 6.2. OpenAI Codex, using GPT-5.6 Sol, investigated, implemented, built, and functionally validated the port through an autonomous coding task.
Towards a Risk Assessment of Malicious Skill Files in Coding Agents (arxiv.org via hn) Autonomous coding agents are increasingly embedded in enterprise software workflows with delegated authority over connected systems. Central to this architecture is the agent skills interface: folders of instructions and scripts that agent…
Show HN: Autonomous Coding, Contained by Default (www.runkiwi.dev via hn) Kiwi runs coding agents inside infrastructure you control, on your own Anthropic, OpenAI or Gemini key. You describe the task and it does that — your own test command is the guard proving the change broke nothing.
Shipwright Harness – open-source autonomous delivery agent for Claude Code (MIT) (github.com via hn) Shipwright Harness The open-source autonomous delivery agent for Claude Code. A deployable cloud agent and the autonomous coding system that powers it — built on the Shipwright plugin, running on your own codebase.
From Prompts to Loops: Building Autonomous Coding Agents (animeshgaitonde.medium.com via hn) Why developers are moving from writing prompts to writing the systems that write prompts 8 min read 1 hour ago Loop Engineering is the new kid on the block. A single tweet in early June has taken the industry by storm.
MiniMax M3 vs. GLM 5.2: Codegen comparison across autonomous coding tasks (thinkwright.ai via hn) Thinkbench, our custom evaluation harness, was used to drive both models through the same autonomous coding loop: read files, write files, run shell commands, and stop when the task was complete. The scored suite covered greenfield builds,…
Show HN: Drydock – VM Sandboxes for macOS Autonomous Coding Agents (github.com via hn) drydock drydock runs autonomous coding agents (Claude Code or OpenAI Codex, per-task selectable) on your own Mac — not someone's cloud — each task sealed in its own hardware-isolated VM. It starts from the assumption that the agent is alre…
Qode – The first AI agent that can generate 50k line codebases in one prompt (github.com via hn) Qode (q-cli) Node SDK v0.1.0 MIT Qode is a terminal-based autonomous coding agent for long-running tasks and massive codebase development. It operates directly in your terminal with a rich TUI, connects to multiple LLM providers, and orche…
Amdahl's law for AI agents (electric.ax via hn) Multi-agent systems are delivering real results. Geoffrey Huntley's Ralph Loop runs autonomous coding agents in a while-loop until every PRD item is complete.
Show HN: Teleport-env – <500ms stateful rollbacks for AI agents via CRIU (github.com via hn) Teleport-Env An ultra-fast, OS-level snapshot and rollback sandbox designed for autonomous coding agents, Monte Carlo Tree Search (MCTS), and reinforcement learning. The Problem Coding agents need environments to test generated bash comman…
Eval/Verifiability for iOS Apps in Claude Code (www.reddit.com) I've been spending time lately on autonomous coding loops for Claude. If the software is easily verifiable, like an API, you can create evals for that and set Claude to build it.
Does it make sense to combine 'everything-claude-code' with RTK (Rust Token Killer) for heavy logging? (www.reddit.com) Hey everyone, I've been tweaking my autonomous coding setup and I'm wondering if this hybrid approach makes sense, or if I'm just engineering a conflict. I use coding agents for complex, multi-file architectures—lately, a lot of Python ser…
The Trillion-Parameter Dilemma: MiMo-V2.5-Pro went open-source (1.02T params). Is self-hosting worth it when the API costs $70 for 387M tokens? (www.reddit.com) Xiaomi open-sourced MiMo-V2.5-Pro. 1.02 trillion parameters, 42B active (MoE), 1M context, MIT license.
ExpertCrowd-SWE: Measuring Frontier Coding Agents on Expert-Authored Engineering Tasks in Unfamiliar Codebases (arxiv.org) ExpertCrowd-SWE (EC-SWE) is a benchmark of 100 expert-authored, multi-file software engineering tasks for evaluating autonomous coding agents in large, unfamiliar codebases. Each task is a change request written by a practising engineer ag…
'Eco Mode' AI coding agent pricing vs token burn. Is running frontier models for test execution a waste? (+benchmark) (www.reddit.com via reddit) recently checked the raw step logs of an autonomous coding agent and it seems the unit economics are pretty absurd to say the least so I watched an agent do what seemed like an ordinary bug fix on a multi-module repo. It resolved it, but t…
"Shall We Approve?" Consequence-Aware Approval for Autonomous Coding Agents (arxiv.org) Coding agents increasingly rely on automated approvers to decide whether a proposed shell command may run. Existing approvers, from rule-based checkers to guardrail models and deployed permission classifiers, base their decisions on the te…
How we solved zombie compilers, runaway retry loops, and transcript disk thrashing in multi-agent coding workflows (www.reddit.com via reddit) Hey everyone, Over the last few months, our team has been running heavy multi-agent workflows using autonomous coding frameworks like Google Antigravity, Cursor, and custom CLI bridges. Moving from single prompt generation to running 3 or…
Policy Search in Code Space for Multi-Agent Pathfinding (arxiv.org) Developing coordination policies for complex multi-agent systems relies on either labor-intensive algorithm design or opaque task-specific reinforcement learning. With recent advances in autonomous coding agents, we show that policy search…
I ran an autonomous coding pipeline for two weeks. It merged 67 PRs, then I shut it down. (www.reddit.comhttps) I built a GitHub Actions workflow where Claude wrote specifications and code, Codex reviewed both, and the agents fixed issues before merging. I controlled the workflow through Telegram.
SpecGuard: Proving a Task Is Broken Before the Agent Cheats (arxiv.org) As autonomous coding agents get increasingly deployed, the risk that accidental or adversarially injected misspecifications in tasks lead to dangerous agent behavior is critical to address. Prior work has shown that agents given such tasks…
Tired of coding agents modifying your unit tests just to fake a "pass"? Here is how to stop them at the runtime level. (www.reddit.com via reddit) Every developer using autonomous coding agents knows this specific frustration: You give the agent a task, a unit test fails, and instead of diagnosing the bug in the implementation, the agent quietly comments out the assertion, slaps .ski…
An Empirical Study of Harness Design for Coding Agents (arxiv.org) Coding harnesses shape how autonomous coding agents translate model capabilities into long-horizon software-engineering performance, yet existing work typically evaluates harnesses as monolithic systems, leaving the effectiveness of indivi…
Assurance Envelopes for Autonomous Coding Agents: Minimum-Cost Evidence for Software Change (arxiv.org) When a coding agent returns to existing software, it inherits evidence from earlier engineering work: tests, type checks, proofs, static analyses, and traces. Reloading all of it is wasteful, but dropping a piece the change depends on can…
I build a payroll saas with CC. Fable xhigh plans, Opus xhigh executes. With the Max 20x usage cut coming I tried putting OpenAI models via Codex CLI into the loop, here are my conclusions. (www.reddit.com via reddit) I've been using CCode since August 2025 and Opus xhigh is my workhorse. I'm a lawyer with payroll domain knowledge but I have some coding background and some good instincts.
TruthInsightBench: An Evidence-Grounded Benchmark for Automated Evaluation of Open-Ended Scientific Discovery Agents (arxiv.org) Autonomous coding agents are increasingly proposed as AI-scientist systems that conduct analyses and write research reports, but executing a prescribed analysis is not the same as making a discovery. Existing benchmarks are configured for…
Can Coding Agents Reproduce Findings in Computational Materials Science? (arxiv.org) Large language models are increasingly deployed as autonomous coding agents and have achieved remarkably strong performance on software engineering benchmarks. However, it is unclear whether such success transfers to computational scientif…
Are highly autonomous coding workflows actually practical without spending hundreds per month? (www.reddit.com via reddit) I've gone pretty deep down the AI coding workflow rabbit hole and I'm curious where people who have tried a lot of this stuff eventually landed. What started as "pick a coding agent" turned into a pretty ridiculous decision tree: Harness:…
Luna high weekly token experience (www.reddit.com via reddit) I am planning to use GPT-5.6 luna high as my main autonomous coding agent. Before this, I was using MiniMax M3, which gives around 1.7B tokens monthly.
I got tired of AI agent loops eating tokens and breaking when VS Code restarted, so I built LoopBoard (turns your TODO.md into a Kanban loop orchestrator) (www.reddit.comhttps) Hey everyone, Like a lot of you running autonomous coding loops directly in VS Code, I kept hitting the same frustrating issues: Lost State: Every time VS Code updated or reloaded, background loops would disappear or get nuked mid-task. To…
Does anyone actually have a fully autonomous coding agent that doesn't need constant follow-ups? (www.reddit.com via reddit) I've been trying to build a fully agentic software development workflow using Claude Code, and I've hit a frustrating problem. The first implementation usually looks good, but every time I ask a follow-up like: "Cross-check everything ag…
NEMO: Execution-Aware Optimization Modeling via Autonomous Coding Agents (arxiv.org) We present NEMO, a system that translates Natural-language descriptions of decision problems into formal Executable Mathematical Optimization implementations using autonomous coding agents (ACAs). Existing approaches rely on specialized la…
Inference Economics of Enterprise Coding Agents: A Case Study of Cloud vs. On-Premise LLMs (arxiv.org) Autonomous coding agents force engineering organizations to choose between API-based frontier models -- strong reasoning at high token cost -- and on-premise quantized open-weights models, which promise low-marginal-cost scaling and data s…
SCATE: Learning to Supervise Coding Agents for Cost-Effective Test Generation (arxiv.org) While autonomous coding agents have significantly advanced automated test generation, they remain fundamentally limited by lazy generation, a phenomenon where agents prematurely terminate tasks and systematically avoid complex programmatic…
Govern the Repository, Not the Agent: Measuring Ecosystem-Level Risk in AI-Native Software (arxiv.org) Autonomous coding agents now open and merge pull requests in shared repositories at scale, and the field evaluates them the way it has always evaluated components, one agent at a time, on isolated benchmark tasks. Yet agents that each pass…
Before the Pull Request: Mining Multi-Agent Coordination (arxiv.org) Autonomous coding agents now open millions of pull requests, yet large-scale studies find their PRs are produced faster but accepted less often - a coordination and trust gap that pull-request-level telemetry cannot explain. We argue the m…
Data Intelligence Agents: Interpreting, Modeling, and Querying Enterprise Data via Autonomous Coding Agents (arxiv.org) Production data integration is bottlenecked by repeated, lossy handoffs between data owners, engineers, and analysts who must collaboratively discover, structure, and query enterprise data. We present Data Intelligence Agents (DIA), a syst…