Hacker News

Show HN: OpenGem – Free, self-healing load-balanced proxy for Google Gemini API

Hacker News - Sat, 02/28/2026 - 4:56pm

Hi HN!

I loved the official Gemini CLI authentication method to get standard Google Gemini AI access, but a single free Google account quota depletes fast.

I wanted to just build and prototype freely. So, I built OpenGem. It essentially turns your idle accounts into your own personal API provider, so developing and scaling side projects is no longer a problem.

GitHub: https://github.com/arifozgun/OpenGem

What it does: OpenGem acts as a standard drop-in replacement endpoint (POST /v1beta/models/{model}). Behind the scenes, it's a smart load balancer. You connect multiple idle/free Google accounts to the dashboard via standard OAuth, and OpenGem routes your traffic to the least-used account.

How it handles limits gracefully: If an account legitimately hits a true 429 quota limit, OpenGem instantly detects it, puts that specific account on a 60-minute cooldown, and seamlessly retries your request with the next available account. The accounts are completely self-healing—a background probe checks them every 30 seconds to see if they've recovered.

It doesn't just blindly retry, either. It uses an 8-category error classifier (50+ regex patterns) to distinguish between a brief rate-limit burst and actual quota exhaustion, and applies exponential backoff with jitter to prevent hammering the servers.

Tech specs:

- 100% compatible with official Google SDKs (@google/genai), LangChain, and standard SSE streaming. - Full support for native "tools" (Function Calling) for agentic workflows. - Raised internal payload limit to 50MB so you can throw huge documents at it. - Has a 3-model fallback chain (flash → pro → pro-3.1) if a specific model is overloaded. - AES-256-GCM encryption for all sensitive configs and OAuth tokens at rest. - Toggle between Firebase Firestore or a fully offline Local JSON database. - It’s completely open-source (MIT licensed) and written in TypeScript. It’s strictly for educational purposes and personal research to bypass the friction of prototyping.

I’m currently running it with my own side projects and it handles agent tasks flawlessly. I would love any feedback on the load balancing and self-healing logic, or just general thoughts!

Comments URL: https://news.ycombinator.com/item?id=47200719

Points: 1

# Comments: 0

Categories: Hacker News

Ask HN: My YC company is hiring one engineer/day but there's not enough work

Hacker News - Sat, 02/28/2026 - 4:56pm

I asked an AI to write this for me so my bosses can’t identify me based on my writing style.

I work at a YC company. It’s not really a startup anymore — we’re well past that phase. We’re hiring at an insane pace right now, roughly one new software engineer per day.

The weird part is… I don’t feel like there’s that much work to do.

I’m not saying there’s no work, but the pace of hiring feels disconnected from reality. I don’t see enough meaningful projects, clear ownership, or actual execution needs to justify this kind of growth. It doesn’t make sense to me.

I feel uncomfortable about it, but I also feel stuck. The job is good. The salary is good. Objectively, I’m lucky. And with the current market — especially with AI changing everything so fast — I’m scared to quit. It doesn’t feel like a safe time to voluntarily jump ship.

After seeing what happened with Bolt and similar companies that rapidly expanded headcount, I can’t shake the feeling that this is just a headcount inflation phase. Grow fast, look bigger, raise more, justify valuations. But with AI accelerating productivity, I keep thinking: what happens when leadership realizes they don’t need this many engineers?

It feels like we’re building up to a correction. And when it comes, it could be brutal. AI tools are already increasing individual output. At some point, won’t companies decide they can do more with fewer people?

I’m honestly lost and scared. I feel like we’re all pretending this makes sense, but deep down it doesn’t. I worry that sooner rather than later, a lot of us are going to be out of a job.

If anyone has advice — whether you’ve been through something similar, or you see this differently — I’d really appreciate it. Right now I feel stuck between staying in something that feels unsustainable and leaving into a market that feels even more uncertain.

Comments URL: https://news.ycombinator.com/item?id=47200708

Points: 1

# Comments: 0

Categories: Hacker News

Show HN: Focusmo – a Mac focus app with a local Claude MCP server

Hacker News - Sat, 02/28/2026 - 4:49pm

I’m the solo developer of Focusmo, a macOS focus app.

I just shipped v7.12, which adds a local MCP server for Claude.

The goal is simple: instead of generic productivity advice, Claude can connect to real focus data and help with weekly reviews, planning, and pattern spotting.

Right now Claude can access: - today’s stats - weekly trends - task list - app usage with an hourly heatmap - personal records - live session state

It can also create tasks and mark tasks complete.

Everything runs locally on Mac through MCP, so the data stays on-device.

I wrote more about it here: https://focusmo.app/blog/claude-ai-mcp-focus-tracking

If anyone from HN wants to try it, I made a code for this thread: CLAUDE50 = $50 off for the first 10 users.

Comments URL: https://news.ycombinator.com/item?id=47200654

Points: 1

# Comments: 0

Categories: Hacker News

A terminal-based KeePass password manager

Hacker News - Sat, 02/28/2026 - 4:40pm

Article URL: https://github.com/shikaan/keydex

Comments URL: https://news.ycombinator.com/item?id=47200579

Points: 1

# Comments: 0

Categories: Hacker News

Reified Generics in PHP

Hacker News - Sat, 02/28/2026 - 4:36pm
Categories: Hacker News

Pages