LATEST TECH ARTICLES

TecAce가 AX Diagnostic을 만든 이유와, 그것으로 무엇을 알 수 있는지에 대하여

Supporting GPU and cloud operations for a 6.5 billion won generative AI project Joins the Konan Technology-led consortium as the specialized GPU infrastructure and cloud MSP partner Provides infrastructure architecture and operational systems for stable execution and operation of a large-scale public AI project Expected to enhance project execution stability by combining Konan's high-performance large language model (LLM) and LLMOps platform with infrastructure and operationa

"You want to put AI on a smartphone?" — The Beginning of a Reckless Challenge The story of how TecAce's AI Supervision Team built the On-Device LLM Tester "Um… I'd like to automatically measure LLM performance on a smartphone." A brief silence fell over the meeting room. Our team was developing our own on-device AI chatbot. The problem was that every time we swapped models, a tester had to physically hold the phone, send prompts one by one, and time everything by hand. That p

DESIGNER IS NOT DEAD In the AX Era, Design Is About Standards, Not Tools Figma or Adobe is not the real issue. In an era where AI tools are becoming part of everyday work, the more important question is this: can your brand and design standards remain consistent even when the tools change? Executive Summary As AI tools continue to grow, questions like “Can we design without Figma?” or “Do we still need Adobe?” have become more common. But these questions miss the bigger point

Gartner says 85% of AI failures come down to data-quality problems. Here are the 3 failure patterns repeatedly seen in SMBs and mid-sized companies in Korea and the U.S., plus a 20-item self-diagnosis checklist to run before adoption. Of these, only 8 truly require humans.

From 12,000 documents scattered across 5 systems to a living knowledge graph of 8,500 nodes and 23,000 relationships. The measured results of six months adopting TecAce AXKH, which combines Ontology-based RAG, a Multi-Agent pipeline, and Human-in-the-Loop governance.

AI must work even where there is no internet. TecAce On-device combines OTA updates with a hybrid offline/online architecture to deliver up-to-date knowledge anywhere — from factory floors out of signal range to campsites deep in the mountains. In this article, we explore five industry use cases where the TecAce platform can be applied, each illustrated with a concrete scenario. Case 1. An AI Manual Companion for Field Workers Field AI Companion — Factories · Shipyards · Plan

Design is no longer only something you make with a mouse. It is becoming something you shape through conversation with AI. Executive Summary: The Future of Design Conversational design tools like Claude Design are revolutionizing how we approach design. Work no longer begins with a blank canvas. Instead, a user describes their needs, and AI generates the first version. This initial output is then refined through conversation, comments, and direct edits. This shift has the pot

Artificial intelligence is evolving fast. One of the most exciting developments is the rise of efficient AI language tools that operate directly on devices. This shift changes how businesses handle data, privacy, and speed. Instead of relying solely on cloud servers, AI can now run locally on smartphones, laptops, or edge devices. This blog dives into the world of on-device large language models, explaining what they are, why they matter, and how they can transform enterprise

You ask ChatGPT to draft an email. You hand Gemini a report to summarize. It feels like having a personal assistant who works only for you. You start to trust it. And that trust is exactly where the problem begins. AI is helpful. That is precisely why it’s dangerous. In 2023, three engineers at Samsung’s semiconductor division pasted source code, internal meeting notes, and hardware design data into ChatGPT three separate times over a single month. They were debugging. They w

DESIGNER IS NOT DEAD From Maker to System Designer Core Message This is not the era where AI replaces design. It is the era where designers define the standards AI must follow. Card Summary As AI creates screens and documents faster, the designer’s role becomes more important, not less. Designers no longer need to make every artifact by hand; they need to design the brand standards and design systems that both people and AI can follow. Executive Summary AI has fundamentally c

Executive Summary AI-powered meeting summaries are nothing new. TecAce Software had already been using various solutions to boost productivity through automated meeting recaps. But two persistent pain points remained: every attendee received the same summary regardless of their role, and recurring meetings lacked the continuity needed to surface meaningful insights. To solve this, the team built an internal Personalized AI Meeting Note System — integrating Speaker Recognition

Executive Summary This case study shares how TecAce Software solved an internal challenge: revenue, cost, and cash flow data scattered across five separate systems — Excel, ERP, SharePoint, and more — made it difficult to get an accurate, real-time picture of the business. As an AI solutions company, TecAce applied its own technology directly to its own operations. Using Vibe Coding for rapid development and a Multi-Agent architecture where specialized AI agents autonomously

Why This Test Matters One SoC generation changed inference speed by 29%. We ran Gemma3 270M INT8 on four Galaxy A-series devices to find out where on-device LLM becomes practically usable. We tested gemma-3-270m-it-int8 via MediaPipe CPU backend on the Galaxy A16, A26, A36, and A56, measuring latency, token throughput, memory, and accuracy across 25 prompts. We also compared parallel (all 4 devices simultaneously) vs serial (each device independently, 2 runs) execution to ver

Click the image to view the report. Every time Google ships a new generation of the Gemma series, the question we ask first is simple: how much faster does it actually run on real hardware? To answer that directly, TecAce ran a head-to-head benchmark between Gemma 3n and Gemma 4 on the Samsung Galaxy S25 Ultra under identical conditions. We tested four model configurations using the llama.cpp CPU inference engine: the previous-generation Gemma 3n E2B Q8_0 as our baseline, a

Galaxy S25 vs S26: On-Device AI Performance Benchmark Reversal! (Snapdragon 8 Elite Gen 1 vs Gen 2) Does a newer chipset always guarantee faster AI performance? Based on real-world test data conducted by TecAce, we compared the on-device LLM performance between the Galaxy S25 and the Galaxy S26 to find out. Test Overview Devices Compared: Galaxy S25 (Snapdragon 8 Elite) vs. Galaxy S26 (Snapdragon 8 Elite Gen 2) Test Models: Gemma3 1B (INT4): An ultra-lightweight conversationa

The Future of On-Device AI and TecAce's Roadmap Throughout this 9-part series, we have chronicled the entire journey of developing an on-device chatbot—a solution to cloud cost and data security issues. We covered everything from selecting a Small Language Model (SLM) and applying quantization, integrating offline STT/TTS, building local RAG, to rigorously validating quality using AI SuperVision and overcoming hardware performance constraints. In this grand finale, Part 10,

Challenging Performance Limits Heat, Battery, and Response Speed In Part 8, we shared how we caught hallucinations and improved response quality using 'AI SuperVision'. While making the model smarter and more accurate is a huge milestone, running it in a real-world smartphone environment (like the Galaxy S25 FE) forces us to confront harsh physical walls: Thermal management, Battery consumption, and Latency limits. Unlike the limitless resources of cloud data centers, a mob

Catching Hallucinations Analyzing SuperVision Test Results In Part 7, we built an automated testing pipeline that bridged our on-device chatbot app inside a smartphone with the AI SuperVision server on a PC. This enabled an end-to-end flow from prompt injection and answer extraction to automated grading. We finally had an environment capable of running dozens of test cases automatically. So, what kind of report card did our on-device SLM (Gemma-2B based) receive from these

Building SuperVision An Automated Chatbot Testing Pipeline In Part 6, we explained the background of introducing Testworks' 'AI SuperVision' tool to objectively evaluate the chronic hallucination issues inherent in generative AI. However, to actually apply this tool to our project, we had to overcome a significant technical barrier. Our LLM chatbot operates completely offline "On-device" (inside a smartphone), whereas the AI SuperVision system evaluating it exists in a "PC