Nvidia profiles Chris Alexiuk, the YouTube teacher making the case that open AI means weights, data and recipes

Open models only democratize AI if someone teaches people how to fine-tune, evaluate and ship them. Chris Alexiuk, a Toronto-based senior product research engineer at Nvidia, has built a YouTube following doing that, and is now the loudest public advocate for Nemotron, Nvidia’s open multimodal model family aimed at long-running agents. He styles himself “No. 1 Nemotron Fan” and “Joe Nemotron” on social media, goes by llm_wizard on X, and describes Nemotron as a spell library: Lightning for high-frequency execution work, Ultra for deep research and enterprise agents. The family has expanded fast, including the 120B-parameter Nemotron 3.5 Super VL with 12B active parameters and a million-token context (Super VL) and small Bolt models for laptops (Nemotron expansion).

“Fully open source AI is necessary for the democratization of intelligence,” Alexiuk says, meaning weights, data, training recipes, everything. That bar is higher than many “open weights” releases clear, and Nemotron’s positioning leans into it. Nvidia’s recent releases have sometimes shipped the data pipeline alongside the model, as with its transcription model and Granary v2 dataset (Nemotron Transcribe), which is the part that lets others adapt rather than merely download. The strategic logic is plain too: generous open models drive demand for the hardware they run on, as European banks building on open weights show (European finance).

His origin story is the best part: a computational physics class, then a TensorFlow image classifier that pinged when a Yellow-Faced Gray Budgerigar entered the webcam frame during co-op games, then a career teaching production LLM apps. The arc from bird alert to laptop-local site builders is the accessibility argument in miniature. He appears at events such as Build-a-Claw at GTC Berlin and keeps up a public teaching practice alongside his job.

The obvious caveat is that the loudest advocate is on the payroll, which invites skepticism of fan accounts tied to an employer’s product line. Alexiuk addresses it by saying he would celebrate being surpassed by someone who found Nemotron through his teaching. The field needs more engineers who shout about open tooling and fewer who treat model cards as marketing; teachers matter more than wizards, and transparent recipes more than hats. More in our Nvidia coverage and research coverage.