TrendPulse

By Hiral Shah, Senior Director, Product Management, Docusign A major recurring theme among the engineering teams at this week’s AI Engineer World’s Fair in San Francisco is the push to move specialized AI models out of research and directly into high-volume production. At Docusign, that optimization challenge happens at massive scale: we handle millions of transactions daily and have nearly 1.9 million customers in over 180 countries. Organizations have historically lost significant valu...

This tutorial was written by Néstor Daza . This is the second article in a series about building Claudius , my own Claude-based chatbot ( Github ). The prologue made the case for building it, and for choosing MongoDB as its foundation. Open the conversations collection in Claudius’ database and you find the usual fields of a thread header but nothing else: a userId , a title , some timestamps , and so on, but no array of messages, no messages collection sitting beside i...

What happens when a machine no longer needs to be trained to see something new? That's the quiet question sitting underneath this week's news, buried next to a less invasive brain implant and a handful of robots getting tougher for the real world. Neuralink says it's completed its first "transdural" brain implant, a surgical approach built to reduce trauma during the procedure. As someone who spends a lot of time thinking about how you get sensors close to a human eye without hurting anyone, ...

by Peter Yang, Behind the Craft Today, I want to share 18 hot takes on where I think the AI market is headed. AI is in a weird place right now. The government is restricting access to frontier models, enterprises are becoming conscious of token costs, and everyone’s trying to rebuild their product for agents first instead of humans. I’ve interviewed dozens of AI leaders and spent far too much time following these topics on X/Twitter. Here are 18 hot takes on where I think AI is heade...

Author: Tobie Morgan Hitchcock Today we're launching Scale, a new tier of SurrealDB Cloud built for the workloads you can't afford to have go down. Our first tier, Start, was designed for building and shipping fast. Scale is designed for what happens next: production traffic, uptime commitments, and the kind of resilience your users never notice because there are always available nodes. It's the tier for teams running SurrealDB as the scalable context layer behind real applications and A...

Nearly half a century ago, the free software movement made a demand that was both technical and moral: Users should have the freedom to understand, run, modify, and share the software on which they depended. It was a demand born from practical life with machines. A printer that couldn't be fixed. A program that couldn't be studied. A system that asked its users to accept dependence as the price of progress. That belief shaped modern computing and gave us the tools and norms that made the i...

You've probably written a CommandExecutor before. Everyone who's touched Bukkit has. Declare the command in plugin.yml , implement onCommand , cast args[0] to whatever you need, hope nobody fat-fingers the input. It compiles. It runs. It's confusing to debug. And it's the wrong way to do it in 2026. # plugin.yml commands : punish : description : Opens the punishment GUI usage : /punish <player> public class PunishCommand im...

Comments

Comments

What are tools or techniques which well for conveying and growing mental model? Techniques I have seen which work well include: Building and maintaining the system (growing) Scribbling on napkins or gesturing next to someone with a furrowed brow (conveying) Documentation which enumerates surface area or catalogs features may be exhaustive but often doesn't achieve either building or conveying the mental model of the subject in question. Using a well designed tool or product directly ...

Comments

Personalized education is moving from theory to practice as AI helps teachers create lessons that fit each student’s needs.

IoT projects often fail not because of weak technology, but because they overlook human behaviour, trust, confidence, and adoption.

The $33B RWA tokenization market is built on a broken data foundation. Traditional PDF inspection reports are designed to dodge liability — not to feed smart contracts. They freeze physical reality at a single point in time and live on-chain forever, silently lying as buildings deteriorate. AI can't fix this: you can't extract facts from a document designed to avoid them. The real solution is a Layer-0 Physical Oracle — a structured JSON payload that translates building defects into hard dollar ...

Entity-based Services and Repositories can add needless complexity when they become thin pass-through layers instead of meaningful abstractions.

Enterprise AI scales only when data, Copilot adoption, agents, security, and platform controls are built as one production foundation.

I see a lot of posts along the lines of AI saved us a ton of money, but they almost never include actual numbers or explain how those numbers were derived. So I decided to break down my approach and be honest about where it gets shaky, because like 10x improvement means nothing without a baseline.

Manufacturing ERP projects rarely fail because of software. Most fail due to unclear goals, inefficient processes, poor data quality, and weak user adoption. Organizations that treat ERP as a business transformation, not an IT project, are far more likely to achieve measurable ROI.

The biggest change AI brings to software is not automation. It is the redistribution of decision making. This article introduces **The Decision Shift**, a product design perspective that explores how AI-native products redefine decision ownership, why trust becomes a core design challenge, and how designers and engineers can build systems where humans remain informed, empowered, and in control.

Video views measure distribution, not understanding. A campaign can generate impressive numbers while attracting the wrong audience or optimizing for weak signals. Better analysis combines retention, audience fit, behavioral patterns, lifecycle stage and downstream actions.

languages. TypeScript consistently generated higher token usage than JavaScript, increasing overall LLM costs by up to 31%.

Modern neural networks are typically explained through optimization, statistics, and linear algebra, which describe how models learn and transform tensors. This article argues that modal logic offers a complementary mathematical framework for interpreting what those transformations represent. Using Layer Normalization, embeddings, attention, residual connections, and hidden representations as examples, it explores how different numerical states can preserve the same semantic structure and how ne...

AI companions and long-term AI assistants need more than intelligence. They need continuity. If users build relationships, projects, memories, preferences, and trust with AI systems, that context should not be trapped inside one platform. The AI Companion Portability Format is an open-source, plain-text way for users to own, edit, move, and preserve their AI context across platforms.

The word cloud creates associations in my mind, and from those associations my imagination begins constructing ideas.

If you're considering PuppeteerSharp for PDF generation, here's the version of the story that doesn't show up in the "getting started" docs.

Title IX was about access. It gave girls and women opportunities that should have always been available. But to me, it also represents something deeper. It helped shift the way society saw what girls and women were allowed to become.

Legacy GPUs remain critical for healthcare, research, and enterprise systems. Here’s how TBR Trade Group helps keep them running.

Brand teams tend to design something once in Canva and then lose hours re-exporting that same layout into a dozen variants for every platform, market, and campaign. In this walkthrough I will show you how to turn a Canva design into a parameterized template, render variations with a single API call, and publish them straight to your social accounts. There is code for engineers and a no-code path through Make and n8n for everyone else.

The AI job apocalypse narrative is fading as real adoption data shows AI reshaping work rather than eliminating entire professions.

Blockchain cannot solve every gaming challenge, but it can improve fairness where trust matters most. This article explores how on-chain records, verifiable randomness, smart contracts, and transparent reward distribution help players verify outcomes instead of simply trusting platforms. It also explains why hybrid Web3 gaming models may offer the best balance between performance, usability, and transparency.