AI / AI Education / AI voices on LinkedIn
The AI researchers and builders most worth following on LinkedIn in 2027 are Andrew Ng for direction, Chip Huyen for production system design, Sebastian Raschka for architectures in code and Maxime Labonne for post-training. Clem Delangue, Jay Alammar, Philipp Schmid, Damien Benveniste and Armand Ruiz complete the list. Each one publishes original technical material, such as a book, course, blog or newsletter, and we confirmed each LinkedIn profile from a link on their own site. LinkedIn blocks automated access, so we describe what each person publishes instead of quoting follower counts we could not see.
Key takeaways
- LinkedIn is a secondary channel for most researchers, so the people here are the ones who publish technical work there in a form engineers can use.
- Split your attention between direction and technique: Ng, Huyen and Raschka for what matters, Labonne, Schmid and Benveniste for how to do it.
- Follow the source behind every post. Each person here links a book, blog or newsletter that carries the full method.
Which AI researchers and builders are worth following on LinkedIn?
Ranked by how much a practitioner or technology leader gets out of each one. "Checked" names the source behind each number and when we looked.
Direction and fundamentals
-
Founder of DeepLearning.AI · 7M+ learners trained (per his site)
For: Engineers at every level who want a reliable signal on which skills and architectures matter next.
Ng led Google Brain at its founding and was Chief Scientist at Baidu before founding DeepLearning.AI and AI Fund. His weekly letters argue for specific engineering practices, such as agentic workflows and fast iteration, and point to new short courses that teach them. For builders, he is the best way to see which skills the field is about to expect.
Checked: andrewng.org About page (career, learner count, LinkedIn link), 19 Nov 2026. LinkedIn blocks automated access, so follower count and posting frequency were not verified.
-
Author of AI Engineering (O'Reilly)
For: Engineers moving LLM prototypes into production systems.
Huyen's own description of her work is bringing AI into production, and her writing focuses on AI system design. Her book AI Engineering is one of the most cited references for building on foundation models. Follow her for structured thinking about evaluation, serving and cost.
Checked: huyenchip.com (bio, LinkedIn link), 19 Nov 2026. LinkedIn blocks automated access, so follower count and posting frequency were not verified.
-
Author of Build a Large Language Model (From Scratch) · LLM research engineer
For: Engineers who want to understand model architectures and training methods in code.
Raschka is an LLM research engineer, formerly a senior engineer at Lightning AI and a statistics professor at the University of Wisconsin-Madison. His book walks through building an LLM from scratch, and his articles explain new architectures and reasoning models. He also appears on our X coverage; we include him here because his architecture explainers are just as useful to readers who live on LinkedIn.
Checked: sebastianraschka.com (bio, book, LinkedIn link), 19 Nov 2026. LinkedIn blocks automated access, so follower count and posting frequency were not verified.
-
Co-founder and CEO of Hugging Face
For: Builders who depend on open-weight models and want to follow where that ecosystem is going.
Delangue leads Hugging Face, the main distribution point for open models and datasets. He is one of the most prominent public advocates for open-source AI. Builders weighing open against closed models get the strongest open-side argument here.
Checked: huggingface.co/clem profile (LinkedIn link), 19 Nov 2026. LinkedIn blocks automated access, so follower count and posting frequency were not verified.
-
Co-author of Hands-On Large Language Models · visual ML explainers
For: Engineers and technical managers who learn best from diagrams.
Alammar is known for visual explanations of machine learning, summed up by his blog's tagline: visualizing machine learning one concept at a time. He co-wrote Hands-On Large Language Models and has moved his writing to Substack. His site names LinkedIn as one of the places he publishes, and his illustrated style makes him the best follow here for visual learners.
Checked: jalammar.github.io (bio, book, LinkedIn link), 19 Nov 2026. LinkedIn blocks automated access, so follower count and posting frequency were not verified.
Hands-on techniques
-
Head of Post-Training at Liquid AI · LLM Course
For: ML engineers who fine-tune and post-train open models.
Labonne leads post-training at Liquid AI and holds a PhD in machine learning from the Polytechnic Institute of Paris. His open-source LLM Course is a common starting point for engineers learning fine-tuning. He is the most technique-dense voice on this list.
Checked: mlabonne.github.io About page (role, education, LinkedIn link), 19 Nov 2026. LinkedIn blocks automated access, so follower count and posting frequency were not verified.
-
Staff Engineer at Google DeepMind
For: Developers building on Gemini who want working code for new features.
Schmid works at Google DeepMind and keeps a technical blog with recent posts on new Gemini models. His writing is code-first, so it suits engineers who want to try something rather than read about it. Expect release-day walkthroughs and agent examples.
Checked: philschmid.de (role, recent posts, LinkedIn link), 19 Nov 2026. LinkedIn blocks automated access, so follower count and posting frequency were not verified.
-
The AiEdge newsletter: tens of thousands of subscribers (per Substack)
For: Engineers preparing for ML system design work or interviews.
Benveniste has a PhD in theoretical physics and was a machine learning tech lead at Meta working on ads ranking. The AiEdge newsletter covers ML engineering techniques, MLOps and deep dives on model families. His diagram-led system-design breakdowns suit LinkedIn's image-first format.
Checked: theaiedge.io About page (bio, subscriber range, LinkedIn link), 19 Nov 2026. LinkedIn blocks automated access, so follower count and posting frequency were not verified.
-
AI with Armand newsletter · Saturdays at 7am ET
For: Engineers and technical leads building AI into business applications.
Ruiz writes AI with Armand, a weekly newsletter of practical tips for putting AI into business use, and founded nocode.ai. The newsletter's About page links his LinkedIn. He sits between builders and buyers, which makes him useful for engineers who also have to explain their work to stakeholders.
Checked: newsletter.armand.so About page (newsletter, cadence, LinkedIn link), 19 Nov 2026. LinkedIn blocks automated access, so follower count and posting frequency were not verified.
Why follow AI builders on LinkedIn instead of only on X?
Many frontier researchers post first on X, and our X coverage reflects that. LinkedIn suits a different kind of technical writer: the explainer who turns a paper or a release into a diagram, a short code walkthrough or a checklist. Alammar, Benveniste and Labonne fit that mold, and their posts reach engineering managers and adjacent roles who are not on X.
LinkedIn is also where hiring managers and team leads spend time. Following builders here keeps you current on the skills employers are starting to expect, which Ng's posts often address directly. If you are planning a team rather than a career, our guide to the AI-native engineering team covers the structure side.
We kept overlap with other lists low. Sebastian Raschka is the one person who also appears on our X coverage; his explainers are essential enough that leaving him out would weaken the list.
What should you not expect from these feeds?
Do not expect breaking research. New papers and model releases usually appear on arXiv, GitHub, Hugging Face and X before anyone summarizes them on LinkedIn. The value of the people here is the explanation that follows a day or a week later, written so that a working engineer can decide whether the change matters to them.
Do not expect neutrality either. Delangue leads Hugging Face, Schmid works at Google DeepMind and Labonne at Liquid AI, so their posts naturally favor their own ecosystems. That is not a reason to skip them; it is a reason to follow more than one of them and to read their claims about competitors with care.
Finally, do not treat engagement as accuracy. Technical posts with clean diagrams spread widely on LinkedIn even when they simplify too far. When a post will change how you build something, check the linked paper, documentation or code before acting on it.
Which builder should you follow first?
Start from your current job. If you ship LLM features, Huyen first, then Schmid for code. If you train or fine-tune models, Labonne first, then Raschka. If you are new to the field or retraining, Ng and Alammar give the best foundation, and Benveniste is useful for system design preparation.
If your work depends on open-weight models, add Delangue. His perspective is that of the company running the largest open-model hub, so read it as advocacy with insider knowledge. Pair it with our comparison of LLM API pricing when you weigh open against hosted models on cost.
Ruiz is the pick for engineers who sit close to business stakeholders. His newsletter is about getting AI into business processes, which helps when you need to translate technical work into outcomes. For the broader list across strategy, marketing and governance, see the LinkedIn AI voices hub, and for an executive filter see CTAIO's executive edition.
Listed here, or should be?
If an entry is out of date or something belongs on this list, email [email protected]. Listings are editorial and free. A featured, labeled Sponsored slot is available on request.
Related questions
- Which AI voices on LinkedIn are worth following across all disciplines?
- Which AI voices should a CTO or CAIO follow on LinkedIn?ctaio.dev
- How should an AI-native engineering team be structured?
- Which AI accounts on X should a tech leader follow?ctaio.dev
Do top AI researchers actually post on LinkedIn?
Some do and many do not. Several frontier lab researchers post mainly on X or not at all. The people on this list publish technical material on LinkedIn or link it there from their own site, which is why we chose them.
Is Yann LeCun worth following on LinkedIn?
He is a major voice, but we could not confirm his LinkedIn profile link from a site he controls, so we left him off this list. We only include profiles we can verify.
Which of these people write books?
Chip Huyen (AI Engineering), Sebastian Raschka (Build a Large Language Model From Scratch), Maxime Labonne (LLM Engineer's Handbook and Hands-On Graph Neural Networks Using Python) and Jay Alammar (Hands-On Large Language Models) have all published technical books.
Ready to Find the Right AI Tools?
Browse our data-driven rankings to find the best AI tools for your team.