Agentic Operations Research
An operations research competition where teams build agentic workflows that design, test and improve solvers for optimisation problems.
Already registered? Sign in
Agentic Operations Research
Instead of hand-tuning a single algorithm, competitors build agentic workflows that design, test and improve solvers for optimisation problems.
Your team builds
Your agent receives the published statement and the instance and solution schemas.
Your agents design, test and improve a solver.
You submit solver code which can solve new problem instances.
Two tracks, scored separately
Open-source dependencies, run in our sandbox
DependenciesOpen-source only. Commercial solvers and remote services are not available.
Runs onOur sandbox. Network disabled, 2 vCPU and 4 GB RAM, a time limit per instance.
You submitSolver code, evaluated against hidden problem instances.
ScoringScored on private instances. The overall champion is selected from this track.
Any solver, run on your own infrastructure
DependenciesAnything. Commercial solvers and proprietary software are permitted.
Runs onYour infrastructure, with results submitted within a time window.
You submitObjective values and complete variable assignments.
ScoringWe verify feasibility and recompute objectives before accepting a result.
One per problem category and one per formulation, each ranked on its own. Sub-track winners will be announced and recognised.
Create an account and form a team to take part in the challenge.
Four weekly rounds of 15 audited problems, 60 in all. Each round is open for 7 days, then closes as the next one opens. Build solvers for each round's problems and climb one leaderboard for the whole stage.
New problems outside the published benchmark, with unseen instances. Details TBA.
Six placed teams share $21,000 in main performance prizes, evaluated on the private final dataset. Two further awards recognise contributions to the benchmark and the best write-up.
$10,000
Winner
First place overall
$5,000
Runner-up
Second place overall
$1,500
Merit Awards
Each, third to sixth place
For novel, high-quality optimisation problems or instance suites that expand the benchmark.
For a thorough, well-documented submission detailing agent methodology, system architecture and findings.
Registration is open
Stage 1 starts 4 January 2027, with 15 new problems each week for 4 weeks. The environment and the entrypoint contract are already in the docs.
Questions? Read the FAQ
