<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: sharathr</title><link>https://news.ycombinator.com/user?id=sharathr</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 18 Nov 2026 04:04:58 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=sharathr" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by sharathr in "Ask HN: What's your current AI agentic setup?"]]></title><description><![CDATA[
<p>try <a href="https://www.droneship.run" rel="nofollow">https://www.droneship.run</a></p>
]]></description><pubDate>Wed, 04 Nov 2026 22:10:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=50372231</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=50372231</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50372231</guid></item><item><title><![CDATA[New comment by sharathr in "Ask HN: What's your current AI agentic setup?"]]></title><description><![CDATA[
<p>have you tried <a href="https://www.droneship.run" rel="nofollow">https://www.droneship.run</a></p>
]]></description><pubDate>Wed, 04 Nov 2026 20:27:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=50371049</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=50371049</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50371049</guid></item><item><title><![CDATA[New comment by sharathr in "Ask HN: What's your current AI agentic setup?"]]></title><description><![CDATA[
<p>try <a href="https://www.droneship.run" rel="nofollow">https://www.droneship.run</a></p>
]]></description><pubDate>Wed, 04 Nov 2026 20:27:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=50371043</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=50371043</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50371043</guid></item><item><title><![CDATA[Show HN: Droneship – sandbox for your coding agents]]></title><description><![CDATA[
<p>We want to share our latest project called Project Droneship (<a href="https://www.droneship.run" rel="nofollow">https://www.droneship.run</a>) that lets you run claude code directly in a cloud sandbox.  You can continue to work on the terminal just like you run normal claude. Droneship also supports Codex, Gemini etc., so a single CLI to run any IDE in a secure sandbox environment. Would love to hear your feedback!</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=50361465">https://news.ycombinator.com/item?id=50361465</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 04 Nov 2026 04:20:56 +0000</pubDate><link>https://droneship.run/</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=50361465</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50361465</guid></item><item><title><![CDATA[New comment by sharathr in "Octomind – open-source AI harness that saves your tokens, static rust"]]></title><description><![CDATA[
<p>have you taken a look at Codeoid: <a href="https://github.com/highflame-ai/codeoid" rel="nofollow">https://github.com/highflame-ai/codeoid</a>, we use it regularly across everyone in our team and its really great.<p>Especially the ability to switch an ongoing conversation from Claude Code to Codex or Gemini (if you run out of tokens).</p>
]]></description><pubDate>Tue, 03 Nov 2026 16:50:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=50354392</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=50354392</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50354392</guid></item><item><title><![CDATA[New comment by sharathr in "Show HN: Ordewell – turn one goal into an ordered plan of coding-agent tasks"]]></title><description><![CDATA[
<p>have you tried <a href="https://github.com/highflame-ai/codeoid" rel="nofollow">https://github.com/highflame-ai/codeoid</a></p>
]]></description><pubDate>Tue, 15 Sep 2026 16:27:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=49714967</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=49714967</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49714967</guid></item><item><title><![CDATA[New comment by sharathr in "If coding has been solved, why does software keep getting worse?"]]></title><description><![CDATA[
<p>we found that using a stable AI-SDLC process like: <a href="https://github.com/highflame-ai/ai-factory" rel="nofollow">https://github.com/highflame-ai/ai-factory</a> improves code agent assisted development dramatically specially when going from 1-100.</p>
]]></description><pubDate>Sat, 25 Jul 2026 02:58:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=49044094</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=49044094</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49044094</guid></item><item><title><![CDATA[Show HN: AI-factory – a spec-driven pipeline that stops AI agents rotting code]]></title><description><![CDATA[
<p>Article URL: <a href="https://github.com/highflame-ai/ai-factory">https://github.com/highflame-ai/ai-factory</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49028457">https://news.ycombinator.com/item?id=49028457</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 23 Jul 2026 21:45:05 +0000</pubDate><link>https://github.com/highflame-ai/ai-factory</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=49028457</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49028457</guid></item><item><title><![CDATA[New comment by sharathr in "How good is your AI Gateway?"]]></title><description><![CDATA[
<p>part one of the blog is here: <a href="https://www.highflame.com/blog/your-ai-gateway-is-fine-until-everyone-hits-it-at-once" rel="nofollow">https://www.highflame.com/blog/your-ai-gateway-is-fine-until...</a></p>
]]></description><pubDate>Thu, 23 Jul 2026 05:07:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=49017184</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=49017184</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49017184</guid></item><item><title><![CDATA[New comment by sharathr in "How good is your AI Gateway?"]]></title><description><![CDATA[
<p>We benchmarked Highflame's AI Gateway against popular alternatives to see how our RUST based unified (LLM/MCP) gateway performs and were surprised at how the competition fared (hint: poorly). We wrote a 2 part blog to review the performance throughput, latency, connection handling in detail.</p>
]]></description><pubDate>Thu, 23 Jul 2026 05:07:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=49017179</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=49017179</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49017179</guid></item><item><title><![CDATA[How good is your AI Gateway?]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.highflame.com/blog/the-three-moments-your-ai-gateway-can-ruin/">https://www.highflame.com/blog/the-three-moments-your-ai-gateway-can-ruin/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49017178">https://news.ycombinator.com/item?id=49017178</a></p>
<p>Points: 3</p>
<p># Comments: 3</p>
]]></description><pubDate>Thu, 23 Jul 2026 05:07:16 +0000</pubDate><link>https://www.highflame.com/blog/the-three-moments-your-ai-gateway-can-ruin/</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=49017178</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49017178</guid></item><item><title><![CDATA[Show HN: Deconstructing Anthropic's Coding Agent Control Model]]></title><description><![CDATA[
<p>Anthropic recently published an excellent write-up on how they contain Claude Code and its sub-agents.<p>One thing that stood out is that the architecture isn’t really about Claude—it describes a general pattern for securing autonomous agents:<p>* Every agent gets its own identity.
* Authority is delegated (and attenuated) to sub-agents instead of being inherited wholesale.
* Tool access is authorized per action.
* Every action is attributable back to the originating principal.<p>We took that architecture apart and mapped it to explore how the same model can be applied across heterogeneous agent fleets (Claude Code, Codex, etc.), not just Anthropic’s own stack.<p>The post focuses on the architecture rather than the product:<a href="https://www.highflame.com/blog/how-anthropic-contains-its-own-coding-agents-and-get-that-coverage-across-your-fleet" rel="nofollow">https://www.highflame.com/blog/how-anthropic-contains-its-ow...</a><p>I’d be interested in feedback from people building agent systems.</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48736437">https://news.ycombinator.com/item?id=48736437</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 30 Jun 2026 17:50:20 +0000</pubDate><link>https://www.highflame.com/blog/how-anthropic-contains-its-own-coding-agents-and-get-that-coverage-across-your-fleet/</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=48736437</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48736437</guid></item><item><title><![CDATA[New comment by sharathr in "[dead]"]]></title><description><![CDATA[
<p>We write about why AI agent sandboxes are insufficient. Identity, not just policy, is critical to securing autonomous AI systems and preventing misuse.</p>
]]></description><pubDate>Tue, 05 May 2026 20:42:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=48028225</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=48028225</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48028225</guid></item><item><title><![CDATA[New comment by sharathr in "Show HN: Agent Vault – Open-source credential proxy and vault for agents"]]></title><description><![CDATA[
<p>From what I can tell, agent-vault does not solve identity, only how its stored. For true agent identity, you should look into: <a href="https://github.com/highflame-ai/zeroid" rel="nofollow">https://github.com/highflame-ai/zeroid</a> (author: full disclosure)</p>
]]></description><pubDate>Thu, 23 Apr 2026 23:06:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=47883390</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=47883390</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47883390</guid></item><item><title><![CDATA[Autonmomous Agent Identity Framework]]></title><description><![CDATA[
<p>Article URL: <a href="https://github.com/highflame-ai/zeroid">https://github.com/highflame-ai/zeroid</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47664112">https://news.ycombinator.com/item?id=47664112</a></p>
<p>Points: 4</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 06 Apr 2026 17:32:37 +0000</pubDate><link>https://github.com/highflame-ai/zeroid</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=47664112</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47664112</guid></item><item><title><![CDATA[New comment by sharathr in "[dead]"]]></title><description><![CDATA[
<p>Research paper here: <a href="https://arxiv.org/pdf/2602.16935" rel="nofollow">https://arxiv.org/pdf/2602.16935</a></p>
]]></description><pubDate>Wed, 25 Feb 2026 19:47:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=47156822</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=47156822</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47156822</guid></item><item><title><![CDATA[New comment by sharathr in "Multi-Turn Intent Detection for LLM and Agent Security (ArXiv)"]]></title><description><![CDATA[
<p>Hi HN — I’m one of the authors.<p>We’ve been working on security for multi-turn agent loops and noticed most detection approaches operate on isolated prompts. This paper introduces a framework for modeling intent trajectories across sequences in real time (<20ms), enabling enforcement before harmful actions occur.<p>Happy to answer technical questions or discuss assumptions in the paper.</p>
]]></description><pubDate>Fri, 20 Feb 2026 21:17:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=47094098</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=47094098</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47094098</guid></item><item><title><![CDATA[Multi-Turn Intent Detection for LLM and Agent Security (ArXiv)]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.arxiv.org/pdf/2602.16935">https://www.arxiv.org/pdf/2602.16935</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47094095">https://news.ycombinator.com/item?id=47094095</a></p>
<p>Points: 1</p>
<p># Comments: 1</p>
]]></description><pubDate>Fri, 20 Feb 2026 21:17:36 +0000</pubDate><link>https://www.arxiv.org/pdf/2602.16935</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=47094095</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47094095</guid></item><item><title><![CDATA[New comment by sharathr in "Palisade: Bringing Zero-Trust to the AI Model Supply Chain"]]></title><description><![CDATA[
<p>We’ve spent decades hardening software supply chains — signing binaries, scanning dependencies, locking down CI/CD — yet AI models themselves are mostly treated as opaque blobs pulled from the internet. That assumption is increasingly unsafe: models can be tampered with, backdoored, or subtly manipulated to behave maliciously at runtime.<p>Highflame’s new tool Palisade brings a zero-trust approach to the AI model supply chain. It validates format and structural integrity, detects hidden malicious patterns, verifies provenance via Sigstore/SLSA, and can even trigger behavioral checks to surface backdoors that only activate under certain inputs. Built in Rust for speed and scalability, Palisade makes it feasible to gate models before they hit inference servers or CI/CD pipelines, turning “download and hope” into a verifiable trust boundary.<p>Author here — happy to answer questions about threat models, performance tradeoffs, or how this fits into CI/CD.</p>
]]></description><pubDate>Wed, 24 Dec 2025 02:12:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=46371719</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=46371719</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46371719</guid></item><item><title><![CDATA[Palisade: Bringing Zero-Trust to the AI Model Supply Chain]]></title><description><![CDATA[
<p>Article URL: <a href="https://highflame.com/blogs/launching-palisade-zero-trust-security-for-the-ai-model-supply-chain">https://highflame.com/blogs/launching-palisade-zero-trust-security-for-the-ai-model-supply-chain</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46371718">https://news.ycombinator.com/item?id=46371718</a></p>
<p>Points: 2</p>
<p># Comments: 1</p>
]]></description><pubDate>Wed, 24 Dec 2025 02:12:48 +0000</pubDate><link>https://highflame.com/blogs/launching-palisade-zero-trust-security-for-the-ai-model-supply-chain</link><dc:creator>sharathr</dc:creator><comments>https://news.ycombinator.com/item?id=46371718</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46371718</guid></item></channel></rss>