Helmcode

Today we’re featured in @LaVanguardia talking about AI, data sovereignty, and why the EU is competing in one of the most important races of our time and how at @helmcode we’re working to help meet that challenge.

🤔 How to reduce LLM and AI infrastructure costs without sacrificing quality We published a guide on cutting AI costs. There's an example in chapter 1 we keep coming back to. You run 10,000 tasks for $330 and 9,000 of them come out right, so each good result costs you about $0.037. Then you move to a cheaper setup: $280 for the same 10,000 tasks, but only 7,000 come out right. Each good result now costs $0.040. You pay less every month and every useful result costs you more. That's why the guide measures everything as cost per accepted task. To save, the first thing that help is splitting the bill by operation. It's 14 chapters guide, and each one can be read as a quick overview or in full detail. There's also a spreadsheet and a SQL query to build your own baseline. helmcode.com/guides/ai-cost-…

🤔 About AI Act: Spain's data protection authority (AEPD) just warned a company that planned to score CVs with AI. The tool wasn't even live yet. The case raises a question many teams building on third-party APIs haven't asked: when does your app make you a high-risk AI provider under the AI Act? You don't need to train a model for that. A system prompt like "score this CV from 1 to 10 and reject anything below 5" is enough to land in Annex III. Put that app into service under your name and you can be its provider. Rodrigo Cilla Ugarte walks through it, using the CV case as an example: → What the AEPD asked for: privacy by design, a DPIA, transparency and human oversight that can actually question the score → Why the CV travelling inside the prompt matters (photo, date of birth, career history) → The three cases in Article 25 → Why "summarise this CV" and "score it and reject below 5" can end up in very different places → Deadlines: Annex III obligations moved to 2 December 2027, but the GDPR and Spain's Royal Decree 723/2026 already apply Plus five things to check before you deploy something like this. Full article here: helmcode.com/posts/high-risk…

When you send a request to an open model on Helmcode, it runs on infrastructure we operate in Europe. We don't forward it to whichever third-party provider happens to be cheapest that day. And we don't store your prompts. It's a fairly simple distinction, but one we think matters a lot when companies start building with real customer data. You should know who's running your AI.

One tweet can start a company. How far can we take it? 🚀 We're launching the Helmcode YouTube channel! And we're kicking things off with this incredible intro created by @Saulgmz_. Absolute cinema. 🚬 Sound on. 🔊 We want to hear what you want to see on the channel. We've prepared a nice form for getting your responses: nan.builders/prompt And if this is just the intro, imagine what's coming next. 😏 youtu.be/c7-QvNDoLvE

This is how our inference architecture works at Helmcode. Serving inference in production is more than hitting the API you use. 🤷‍♂️ You need routing, fallbacks, observability, cost control, and the right deployment model. Here's a closer look at how we approach it at Helmcode: 1/ API Gateway & Routing Every request enters through an OpenAI-compatible API. We handle authentication, rate limits, and routing to the appropriate model and infrastructure. 2/ Model Serving Different models have different requirements. We run inference workloads across GPU infrastructure, managing model replicas, batching, and concurrency to balance throughput and latency. 3/ Scalability & Reliability Traffic isn't constant. The infrastructure needs to handle demand spikes, distribute requests across available capacity, and recover when something fails. 4/ Observability We monitor latency, token throughput, GPU utilization, errors, and request volume to understand how the system is performing and identify bottlenecks. 5/ Cost & Deployment Infrastructure decisions directly affect inference costs. Shared, dedicated, and on-prem deployments have different trade-offs in utilization, isolation, performance, and economics. All of this needs to work together while keeping the API simple for developers. This is how we make open model inference accessible to businesses at Helmcode.

💸 How to fix your AI bill. A comprehensive guide about AI cost optimization explained by our platform team 👇 We've just published a practical guide to reduce your AI cost. 14 chapters covering model selection, routing, caching, agent loops, RAG, GPU economics, and the decisions that actually reduce inference costs. It's free, with three reading depths, real calculations, a capacity calculator, a spreadsheet, SQL templates, and a downloadable PDF. 👉 Read the full guide: helmcode.com/guides/ai-cost-… The quick summary, in 5 points: 1/ Stop optimizing for price per token. A model that's 5x cheaper means little if it needs three attempts to complete the same task. Measure cost per successful task, including retries, tool calls, and failures. 2/ Your best model shouldn't handle every request. Classification, extraction, summarization, coding, and complex reasoning have different requirements. Route tasks to the models that can handle them reliably. Save expensive reasoning for the work that needs it. 3/ You're probably paying for the same work more than once. Repeated context, duplicate tool calls, unnecessary retrieval, and agents that keep searching after finding the answer. Caching, context management, and better stopping rules can make a real difference. 4/ Agents need budgets, not just prompts. An agent can turn a cheap request into an expensive workflow through retries, searches, and tool calls. Set limits on steps, time, and spending. Track the cost of the complete workflow. 5/ Self-hosting isn't automatically cheaper. A GPU sitting idle still costs money. Compare infrastructure, utilization, peak demand, latency, and engineering time against API pricing. Sometimes dedicated infrastructure wins. Sometimes pay-per-use does. Sometimes a flat rate makes more sense. We've also included a fictional company that starts with a $2,600/month AI bill. Across the guide, we walk through the decisions that change that bill, one by one. No magic 90% savings claim. Just the calculations and trade-offs behind each decision. Free guide, available in EN/ES: helmcode.com/guides/ai-cost-…

The @Revolut thing is getting noticed... How their engineering team cut their AI bill by 90%. But a lot of people/ companies are wondering how they moved to open models. Here’s a short video on how they did it and how you can do it too.

DeepSeek is open-sourcing their full AI ecosystem: models, harness, inference and now, agents. We’ve been testing DeepSeek Super-agent to review some PRs, and here’s a full breakdown of what it is, how it’s built and how it works 👇

Just use helmcode.com 🤷‍♂️ Cut your AI bill with open models flat rate pricing as @Revolut did 👇 You don't need to self-host and pay for a bunch of h200... You can use our API to call the best open models: - GLM 5.3 - DeepSeek v4.1 Flash - Qwen 3.6 & Gemma unlimited And with zero logs and with server on EU ✌️

👀 @Revolut has reduced its AI coding costs by 90% by self-hosting open models. And I think their story is worth paying attention to. At @aiDotEngineer NYC, David Abitbol shared how Revolut went from being skeptical about the ROI of closed models to making open-weight models a priority for its engineering teams. Some interesting numbers from their experience: - Around 2,000 engineers using AI tools. - 90% lower AI costs compared with their previous Anthropic and OpenAI spending. - 60% improvement in coding time. - 70% improvement in lead time. - GLM models running on their own GPU infrastructure. 👉They still use closed models for tasks where those models perform better. But open models have become their default for much of their coding workload. One of the most interesting points from the talk is what happens when engineers no longer have to worry about token budgets. They start using AI for more of their work, including the tasks surrounding coding that often slow down delivery. ❤️ This is very close to what we believe at Helmcode. We're seeing more companies reconsider their dependence on closed-model APIs, especially as usage grows and costs become harder to predict. But there's an important difference between Revolut and most companies. Revolut has the scale, engineering resources, and infrastructure to operate its own models. Many teams don't. That's why we built Helmcode around operating open models for companies that want the benefits without having to manage GPUs, inference servers, and deployment themselves. We also offer flat-rate pricing, so teams can scale usage with a predictable bill. We've already seen customers reduce their AI costs by up to 75%. And while there will always be workloads where frontier closed models are the better choice, we're increasingly seeing open models become a practical default for production applications. Revolut's results are a great example of where this is heading. Really worth watching David's talk. It's technical, practical, and full of useful numbers. youtube.com/watch?v=Ok6QdZHC…