Hacker News

Subscribe to Hacker News feed
Hacker News RSS
Updated: 3 min 17 sec ago

How to train your CNN model?

Tue, 03/25/2025 - 3:54am

I am currently implementing and training some CNN with a UNet Architecture, therefore I need to train my model for several hours.

I got Colab Pro, but not Colab Pro+, although I need to leave my browser and let the model train in the background.

Which services are the best to do such?

I was looking at runpod.io any other suggestions?

Comments URL: https://news.ycombinator.com/item?id=43468890

Points: 1

# Comments: 0

Categories: Hacker News

AI-Driven Hiring: How MokaHR Transforms Talent Acquisition

Tue, 03/25/2025 - 3:47am

Hiring isn’t just about filling positions—it’s about building high-performing teams. At MokaHR, we leverage AI to streamline recruitment, ensuring companies find candidates who aren’t just qualified but truly aligned with their culture and goals.

Intelligent Matching: Precision in Hiring

Traditional hiring processes are inefficient and biased. MokaHR’s intelligent matching function uses advanced algorithms to analyze multiple data points, pairing candidates with roles where they can thrive. This reduces guesswork and ensures quality hires that contribute to long-term business success.

Data-Driven Hiring Decisions

Gut-feeling hiring leads to costly mistakes. MokaHR provides real-time, data-rich insights, helping recruiters make evidence-based decisions. Our AI examines candidate backgrounds, competencies, and career trajectories, creating comprehensive profiles that go beyond surface-level resume details.

AI-Powered Candidate Profiling

MokaHR doesn’t just scan resumes—it understands candidates. Our system evaluates professional history, skill relevance, and key performance indicators, ensuring hiring managers get a clear picture of each applicant’s potential. The result? More accurate, bias-free hiring.

The Business Impact

Hiring better talent faster isn’t just a convenience—it’s a competitive advantage. Companies using MokaHR experience reduced time-to-hire, stronger teams, and higher retention rates. Our AI-driven approach ensures that every hire adds measurable value, fueling growth and innovation.

The Future of Hiring

The hiring landscape is shifting, and companies still relying on manual processes will fall behind. AI-powered recruitment isn’t a luxury; it’s a necessity. MokaHR is at the forefront of this transformation, redefining how businesses attract, evaluate, and secure top talent.

Want to see how AI can optimize your hiring process? Let’s talk.

Comments URL: https://news.ycombinator.com/item?id=43468865

Points: 1

# Comments: 0

Categories: Hacker News

Building SubTrackr – A Simple Subscription Tracker for Small Businesses

Tue, 03/25/2025 - 3:41am

Hey HN!

I’m currently building SubTrackr, a lightweight subscription tracking SaaS designed for small businesses. The idea is to help companies monitor, manage, and optimize their recurring expenses by connecting bank accounts, categorizing subscriptions, and providing actionable insights.

A bit about me: I’ve been diving into entrepreneurship and SaaS development, and this is my first SaaS project. I run an e-commerce business (Onyx Vault), have experience with dropshipping and finance, and I’m now looking to build a simple but useful tool that solves a real problem.

Unlike Rocket Money, which focuses on individuals, SubTrackr is tailored for businesses—helping them track software subscriptions, SaaS expenses, and recurring costs with automated reminders and integrations.

Would love to get your feedback! What pain points do you have when managing subscriptions for your business? What features would make a tool like this essential for you?

Excited to hear your thoughts!

Comments URL: https://news.ycombinator.com/item?id=43468845

Points: 1

# Comments: 0

Categories: Hacker News

Scaling Hiring with AI: How MokaHR Automates Recruitment

Tue, 03/25/2025 - 3:41am

Hiring is broken. Many companies still rely on outdated processes—manual resume screening, endless email chains, and chaotic interview scheduling. At MokaHR, we believe recruitment should be as streamlined as modern software development. That’s why we built Moka Recruiting, an AI-powered applicant management system designed to automate hiring and eliminate inefficiencies.

Why an Applicant Management System Matters In a hyper-competitive job market, speed and accuracy in hiring are critical. The cost of a bad hire is high, but so is the opportunity cost of a slow process. Moka Recruiting automates the most time-consuming parts of recruitment—resume screening, interview scheduling, and applicant tracking—so HR teams can focus on hiring great talent instead of wrestling with logistics.

How AI Improves Hiring Automated Screening: Our AI scans and ranks resumes in real time, ensuring top candidates don’t slip through the cracks.

Intelligent Scheduling: No more back-and-forth emails—our system syncs with calendars to auto-schedule interviews.

Data-Driven Decision Making: Built-in analytics help track hiring trends, measure conversion rates, and optimize recruitment strategies based on real-time insights.

Enhanced Candidate Experience: Automated communication keeps applicants in the loop, reducing drop-off rates and improving employer branding.

Why MokaHR? MokaHR isn’t just another ATS (Applicant Tracking System). We leverage cutting-edge AI to make hiring not just easier but significantly more efficient. Companies using Moka Recruiting see reduced time-to-hire, lower hiring costs, and improved candidate quality.

The Future of Hiring Recruitment shouldn’t be stuck in the past. If your company is still manually screening resumes and struggling with scheduling, it’s time to rethink your process. AI-powered hiring is no longer a luxury—it’s a necessity.

Curious about how MokaHR can optimize your hiring process? Let’s talk.

Comments URL: https://news.ycombinator.com/item?id=43468840

Points: 1

# Comments: 0

Categories: Hacker News

Ask HN: What LLM tools to locally search and analyse your documents?

Tue, 03/25/2025 - 2:42am

I have yet to find any easy to use local models that I can feed my desktop archive of files and folders and then give it queries on that material. Are there any usable alternatives?

Many years ago I tried Devonthink, before LLMs were a thing, but it locked you in an app and was not so powerful.

Comments URL: https://news.ycombinator.com/item?id=43468591

Points: 1

# Comments: 0

Categories: Hacker News

Pages