https://www.remoroo.com

>REMOROO
BenchmarksPricingDocsEngineering Journal
Install CLIBook a demoPyPIStar on GitHubSign inGet started
Toggle menu

Autonomous engineeringfor deep tech teams

Remoroo runs autonomous research on your code locally, overnight. It
edits, tests, evaluates, keeps or reverts. You wake up to better
results.

$pip install remorooRead the docs -
> remoroo-session * autoresearch/mar31

> Reading program.md... 
> Baseline: val_bpb = 2.2396 (commit 9138841)
> Time budget: 20 min per experiment
> 30 experiments * 8 kept * 22 discarded
> val_bpb: 2.2396 - 1.5484 (31% lower)
> Verdict: VERIFIED * REPRODUCIBLE

The reality of manual ML research

Without Remoroo
$ vim train.py
> tweak learning_rate=3e-4
$ uv run train.py
> wait 60 minutes...
> val_bpb: 2.24 (no change)
> try batch_size 2^15...
> wait 60 more minutes...
> NaN loss.
$ git checkout .
2 hours. 0 progress.
no verdict. no structure.
no proof.
With Remoroo
$ remoroo run --local
program.md
> 30 experiments completed
> 8 kept * 22 discarded
> val_bpb: 2.24 - 1.55
> VERIFIED * REPRODUCIBLE
You slept through it.

How it works

Write a spec (e.g. program.md). Point Remoroo at it, and it runs
experiments overnight.

> remoroo-session * autoresearch

Spec program.md (TIME_BUDGET=1200, metric: val_bpb)
File train.py (model, optimizer, training loop)
Eval prepare.py - evaluate_bpb (fixed, untouchable)
P
Plan
E
Edit
T
Train
E
Evaluate
val_bpb
--
vs baseline
--
train.py
- ATTN_PATTERN = "L" * DEPTH
+ ATTN_PATTERN = "SSSL"
Illustrative billing * credits = Haiku-hour units (x model tier -- see
Pricing)

Verified results

LR SCHEDULE SEARCH
val_bpb
2.24 - 1.99
11% lower
train.py
14 experiments * 6 kept
VERIFIED
ARCHITECTURE SEARCH
val_bpb
1.55 - 1.55
banded attn (SSSL)
train.py
30 experiments * 8 kept
VERIFIED
MULTI-OBJECTIVE
val_bpb + memory
3 constraints - all passed
all passed
train.py
22 experiments * 5 kept
VERIFIED
Explore all benchmarks -

Not a coding agent.

An autonomous research engine.

                Coding Agents                 Remoroo
Time scale      Seconds       Hours to overnight
Task scope      Fix one bug   30-experiment search
Execution       None /        Sandboxed, time-budgeted
                one-shot
Metric          None          Fixed eval harness
evaluation
Keep / discard  Human decides Autonomous, metric-based
Failure         Retry prompt  Case-based recovery
handling
Output          Suggested     Verified patch + proof
                code
Reproducibility None          Artifact replay + git
Billing         Per token/    Run wall time in credits (Haiku-hour
                seat          units)

It didn't guess. It proved.

Install in 30 seconds.

Free tier includes monthly run credits -- see Pricing.

$pip install remoroo
Install CLIRead the docs -
>REMOROO(c) 2026 Remoroo
PricingDocsGitHubTermsPrivacy[email protected]