Twolabs

Active · RL environments for underexplored domains.

YC X26

Website YC profile

Why it ranks

How the rank works

7signal of 100 · rising 7

went through YCProgram · this site+4
team of 2Team size · YC's directory+2.5

A sum of public facts, not a judgement and not investment advice.

Competitors

hillclimbShared: recursive self, self improvement, training dataYC F25San FranciscoStartup →AI

Cashew LabsShared: recursive self, self improvement, self improvingYC X25San FranciscoStartup →Developer tools

Preference ModelShared: learning environment, reinforcement learningSPC 2025San FranciscoStartup →AI

CartpoleShared: learning environment, reinforcement learning, trainingYC X25San FranciscoStartup →AI

SCENARIOXShared: model fall, fall short500 Global 2025Saudi ArabiaStartup →Other

Prime IntellectShared: reinforcement learning, evaluationStartup →Other

Patronus AIShared: environment, training, evaluationStartup →

Companies closest to Twolabs
Company	Category	Status	Program
hillclimbShared: recursive self, self improvement, training dataYC F25ActiveAI, Devtools	AI, Devtools	Active	YCF25
Cashew LabsShared: recursive self, self improvement, self improvingYC X25ActiveAI, Devtools	AI, Devtools	Active	YCX25
Preference ModelShared: learning environment, reinforcement learningSPC 2025Not statedDevtools	Devtools	Not stated	SPC2025
CartpoleShared: learning environment, reinforcement learning, trainingYC X25ActiveAI, Data	AI, Data	Active	YCX25
SCENARIOXShared: model fall, fall short500 Global 2025Not statedAI, Data	AI, Data	Not stated	500 Global2025
Prime Intellectnot in a program hereShared: reinforcement learning, evaluationElsewhereOther	Other	Elsewhere
Patronus AInot in a program hereShared: environment, training, evaluationElsewhereOther	Other	Elsewhere

Adjacent

Zettascale ComputingShared: recursive self, self improvement, chipYC S24San FranciscoStartup →AI

JigsawShared: recursive self, self improvement, self improvingspeedrun SR005San FranciscoStartup →Developer tools

FernShared: robot policy, robotics, environmentYC W26Startup →AI

AviroShared: training dataset, learning environment, reinforcement learningYC X25San FranciscoStartup →AI

HubShared: training data, robotics, frontierYC X26San FranciscoStartup →

Companies next to Twolabs
Company	Category	Status	Program
Zettascale ComputingShared: recursive self, self improvement, chipYC S24ActiveAI, Devtools	AI, Devtools	Active	YCS24
JigsawShared: recursive self, self improvement, self improvingspeedrun SR005Not statedAI, Data	AI, Data	Not stated	speedrunSR005
FernShared: robot policy, robotics, environmentYC W26ActiveDevtools	Devtools	Active	YCW26
AviroShared: training dataset, learning environment, reinforcement learningYC X25ActiveAI	AI	Active	YCX25
HubShared: training data, robotics, frontierYC X26ActiveAI, Data, Devtools	AI, Data, Devtools	Active	YCX26

Also from X26 in AI

All of X26

Advanced Metal ResearchMachine intelligence for American WeldingYC X26Los AngelesStartup →AI

General InstinctDeployment infrastructure for Physical AIYC X26San FranciscoStartup →AI

InLoop RoboticsRobotic arms that pick and pack your orders.YC X26BYStartup →AI

PerfectBit, Inc.RoboticsYC X26San FranciscoStartup →AI

StableBrowseBuilding the Data Infrastructure layer for Physical AI.YC X26San FranciscoStartup →AI

AbInitio BioThe Intelligence layer for Drug ManufacturingYC X26BostonStartup →

Company	Category	Status
Advanced Metal ResearchMachine intelligence for American Weldingteam of 3ActiveAI	AI	Active
General InstinctDeployment infrastructure for Physical AIteam of 2ActiveAI	AI	Active
InLoop RoboticsRobotic arms that pick and pack your orders.team of 10hiringActiveAI	AI	Active
PerfectBit, Inc.Roboticsteam of 3hiringActiveAI	AI	Active
StableBrowseBuilding the Data Infrastructure layer for Physical AI.team of 3ActiveAI	AI	Active
AbInitio BioThe Intelligence layer for Drug Manufacturingteam of 2ActiveAI	AI	Active

Founders

Sardor RahmatulloevFounder/CEO
Danyal AhmadFounder/CTO

About Twolabs

We build verifiable reinforcement-learning environments, evaluations, and proprietary training datasets to improve frontier AI models and robot policies. We focus on complex, underexplored domains where high-quality training data is scarce and existing models fall short. Our long-term goal is to develop self-improving training infrastructure that can autonomously generate increasingly challenging tasks, evaluate performance, and continuously advance model capabilities toward recursive self-improvement.

YC batch: X26 · Spring 2026
Status, per YC: Active
Category: AI, Developer tools
Team size: 2
Location: San Francisco, CA, USA
Founded: 2026
Checked: 13h ago

YC tags: Robotics, AI