In this edition of our #EmployeeSpotlight, we sit down with Cezar Derevlean, one of AscentCore’s talented Technical Leads. We explored his journey from Senior Software
Stories
Get to know AscentCore through the eyes of our talented team members.
Discover how Andreea Rusu brings together communication, respect, and a bit of creativity to build strong teams and drive success as an Engineering Manager at AscentCore.
Meet Ana, our dynamic Software Engineer, who's not just all about coding but also enjoys hitting the gym, biking through forests, and bonding over volleyball matches. From her love for fitness to her dedication to software development, Ana embodies the perfect blend of passion, talent, and zest for life. Let's get to know more about her journey and what makes her tick. Ready? Here we go!
Curious about what makes Norbert Csornai tick? In this interview, our Engineering Manager shares his passion for football, motorcycle rides, and exploring new places with his wife. He opens up about his career journey, his unique leadership style, and why building relationships is key at AscentCore. Norbert also talks about his love for tech, his favorite gadgets, and the lessons he's learned along the way. Don't miss his insights on what's next in his career and the skills he's eager to develop, especially in AI.
You may have seen his picture recently shared all over our socials, as he was one of the speakers at our latest AscentCore Tech Communities Meetup. Ladies, gents, we're proud to present Robert Preda. So get ready to take a sneak peek into the life of Robert. From gym sweat sessions to late-night gaming escapades, his world is a blend of fitness, fun, and code-crunching adventures. Join us as we uncover the story of how this tech-savvy gamer turned into a Senior Software Engineer at AscentCore. Let's see what Robert is all about when he's not crushing it as a speaker - and about that, he did prepare a complete set of tips and tricks to being a great presenter. Do NOT miss out!
In today's Employee Story, we explore Zoli's professional journey, our esteemed Technical Lead (even though he can't believe this is his actual job title - you'll get this in a few minutes), as he offers insights into his experiences navigating the tech landscape and life at AscentCore. In this interview, Zoli provided us with some valuable perspectives on his career evolution, highlighting the challenges and rewards of working in the tech industry. From his early days experimenting with computers to his current role shaping software solutions, Zoli's narrative provides a window into the day-to-day realities of a tech professional. And Zoli really is a professional, in every sense of the word.
Large language models respond to emotionally charged inputs with contextually appropriate outputs, but the mechanism by which they represent, propagate, and modulate emotional tone through their internal layers remains poorly understood. Do emotions "live" in specific layers? Is the signal carried by the attention mechanism, the MLP, or the residual stream itself? And when a model is instructed to be a "helpful assistant," does its internal representation remain emotionally neutral, or does it mirror the user's emotional state?
The tool is never the bottleneck. The bottleneck is everything the tool cannot see, cannot access, and does not know it should ask about. The
An organisation that cannot remember what it decided, or why, is condemned to decide the same things over and over again, each time believing it is the first.
This report presents the results of a systematic evaluation of 22 quantized open-source language models across description generation tasks, measuring quality, JSON reliability, and inference efficiency.
Your best feature may be destroying your margins, and your engineering team has no idea. This article isn’t about AI as a productivity tool.
“The information you never see is, by definition, the information you cannot evaluate. And when the machine decides what is relevant on your behalf,
Large language models respond to emotionally charged inputs with contextually appropriate outputs, but the mechanism by which they represent, propagate, and modulate emotional tone through their internal layers remains poorly understood. Do emotions "live" in specific layers? Is the signal carried by the attention mechanism, the MLP, or the residual stream itself? And when a model is instructed to be a "helpful assistant," does its internal representation remain emotionally neutral, or does it mirror the user's emotional state?
The tool is never the bottleneck. The bottleneck is everything the tool cannot see, cannot access, and does not know it should ask about. The
An organisation that cannot remember what it decided, or why, is condemned to decide the same things over and over again, each time believing it is the first.
This report presents the results of a systematic evaluation of 22 quantized open-source language models across description generation tasks, measuring quality, JSON reliability, and inference efficiency.
Your best feature may be destroying your margins, and your engineering team has no idea. This article isn’t about AI as a productivity tool.
“The information you never see is, by definition, the information you cannot evaluate. And when the machine decides what is relevant on your behalf,