November 13 to 15, 2026 · Chicago

CIQ Supercomputing Local AI Hackathon

One weekend, two GB10 nodes per team, and 14 projects built to use the whole cluster.

A build weekend ahead of SC26

The idea started at ISC 2026 in Hamburg, where CIQ gave away a DGX Spark and asked attendees what they would build on it. Seventy-three people answered with real project ideas, so we set up a weekend to build some.

The weekend before Supercomputing 2026, teams met at the Drive Capital office in the Fulton Market District of Chicago for a hands-on AI build weekend. Each team had two Lenovo ThinkStation PGX machines, powered by the NVIDIA GB10 Grace Blackwell Superchip, connected into a multi-node environment running RLC Pro AI and orchestrated through Fuzzball.

The challenge: build a real AI workload that takes advantage of the full cluster, and show by Sunday what the second node added.

The 14 submissions are collected below. Each page covers what the team built, how they used the cluster and what they measured, in their own numbers.

Challenge tracks

Teams picked one of four tracks to build toward.

  • Track 1

    Multi-Node Training

    A training or fine-tuning workload that takes advantage of compute across multiple nodes.

  • Track 2

    Inference at Scale

    Multi-node inference for greater throughput, concurrency, model capacity or performance.

  • Track 3

    Scientific AI

    Distributed AI and accelerated computing applied to a scientific or research workload.

  • Track 4

    Cluster & Scheduling

    Work on orchestration, scheduling, resource allocation, utilization or multi-node infrastructure.

Scenes from the weekend

  • Participants seated on tiered benches watch a presentation on the NVIDIA Agent Toolkit.
    A kickoff talk on the NVIDIA Agent Toolkit.
  • A speaker addresses the room in front of the hackathon banner and projector screen.
    Opening remarks.
  • A mentor raises a hand while the host introduces the people on hand to help teams.
    Meeting the mentors.
  • Four people stand in front of an NVIDIA banner while one of them speaks to the room.
    A welcome from the sponsors.
  • A group works on laptops around a conference table, with a diagram on a whiteboard behind them.
    The Team 15 room on Saturday morning.
  • A participant explains a surrogate model diagram drawn on a whiteboard.
    Sketching a surrogate model: physics on one node, a neural network on the other.
  • Two people lean over a laptop at the Team 3 table in a glass-walled room.
    Working through a problem at the Team 3 table.
  • A portable monitor shows a terminal session beside energy drinks and laptops.
    Terminals and caffeine.
  • Six people work on laptops on sofas and armchairs around a coffee table.
    Teams spread out across the venue.
  • Three people smile at the Team 5 table beside an open laptop.
    At the Team 5 table.
  • Organizers and mentors talk in a meeting room.
    Organizers and mentors between sessions.

Clips from the weekend

  • Kickoff: build starts now

    Kickoff: build starts now

  • One weekend, a GB10 cluster per team

    One weekend, a GB10 cluster per team

  • Hacking is over. Show what you built.

    Hacking is over. Show what you built.

The projects

Training, inference, scientific AI and scheduling work, all run on two nodes.

  • Team 1

    CoopMoE

    Serves a 235B mixture-of-experts model that does not fit in the memory of both machines combined, by letting each node borrow experts from the other before it reads its own disk.

    Inference at ScaleView project

  • Team 2

    Split-Precision KV-Cache LLM Serving Engine

    A serving gateway that treats KV-cache precision as a scheduling choice, sending each session to the BF16 or FP8 worker with the shortest predicted completion time.

    View project

  • Team 3

    Alchemy

    An agent that runs recipe, training and inference experiments on your own hardware and hands back one Fuzzball workflow that fine-tunes a model and serves it.

    Multi-Node TrainingView project

  • Team 4

    Care Plan Check

    Reviews a clinician’s care plan against the visit, the chart and public local data, with the patient side and the web side running as separate Fuzzball workflows.

    View project

  • Team 5

    Project Prometheus

    Ensemble inference that maps where an ink-detection model can be trusted on an unopened Herculaneum scroll, run as a Fuzzball task array across two nodes.

    View project

  • Team 6

    Honest Harvest

    Runs training experiments on the idle time of the GPU that hosts an agent’s LLM, and pauses them at the next training step whenever an agent needs the model.

    View project

  • Team 7 (CLIF)

    ICU Horizon

    Five 24-hour predictions for every patient in an intensive care unit, summed into what the unit will need, from a model trained and run inside the hospital.

    View project

  • Team 8

    Alder Audit

    Helps Chicago residents find City Council conversations, follow the issues they care about and see what was said, with quotes checked against meeting transcripts.

    View project

  • Team 9

    FinScale

    Fine-tunes NVIDIA Nemotron-3-Nano 30B on two nodes to turn long annual reports into structured risk signals linked to the source text.

    View project

  • Team 11

    Surgical Vision

    A surgical review demo from Serelora and OpenSurgery AI that ties a depth orbit, a source-based report and cited clinical reasoning to one frame, and reuses model outputs between views.

    View project

  • Team 12

    PRAVAH

    A flood intelligence prototype for Nepal that combines satellite water mapping, exposure screening and a map dashboard, with U-Net training distributed across two GB10 nodes.

    View project

  • Team 14

    wxrouter

    A global weather model trained from scratch, served behind one address across two GB10 machines so forecasts keep flowing while a machine fine-tunes or a server dies.

    View project

  • Team 15 (HPC Hacks)

    Ember

    Trains a U-Net to imitate the ForeFire wildfire simulator, with CPU cores generating new simulations while both GPUs train one shared model.

    View project

  • Team 15 (solo entry)

    LymeScope

    Research software that makes every score about a Lyme disease surface protein traceable to its observations, its experimental structure and a reproducible calculation.

    Scientific AIView project

Where we plan to hack next

We plan to run hackathons at these events. Dates and details will follow.

  • March 2027

    NVIDIA GTC

    San Jose, California

  • November 2027

    SC27

    Denver, Colorado

Sponsor a future hackathon

Put your hardware, platform or team in front of the people building on it. Leave your details and our events team will follow up.