Mamori raises $5M to build ML-enabled Pathfinder for blockchain value extraction, led by Blockchain Capital
Share Article
Article Outline
Mamori, an automated blockchain value extraction system, has raised $5 million in a seed funding round led by Blockchain Capital, with participation from Velocity Capital and Web3.com. Mamori is pioneering a general-purpose pathfinder for detecting and defending against “unknown unknowns”, namely zero-day vulnerabilities in smart contracts and new blockchain-related software.
"_"We couldn't be more excited to support Mamori in their vision to make smart contracts more secure via modern advanced fuzzing techniques. By enhancing the developer toolkit, this will enable devs to spot more vulnerabilities earlier in a more scalable way." "
Angel investors include Grigore Rosu, Daniel Lubarov, Alex Watts, Bo Du, Shujia Liang, Micheal Heinrich, Richard Adjei and Antonio Viggiano.
What is Mamori?
Mamori is an automated value extraction system optimized for blockchain security and value extraction, colloquially named a ML-enabled pathfinder.
Cyber attacks remain a pestilent issue in crypto, with over US$2.4 billion in losses attributed to exploits in 2023. Existing auditing tools and methodologies, derived from historical data, are ill-equipped to meet the preventive security needs of an industry with rapid innovation and billions of dollars at stake.
"_“We are very excited about Mamori’ endeavor to identify zero-day exploits by ML-enabled fuzzing to test continuous function input parameters and function sequences. This way, they make smart contracts more secure and more institutions repose confidence in the space, thus facilitating more capital inflow to the on-chain economy, an essential step towards mass adoption!”"
Mamori disrupts web3 security with a versatile, future-proof offensive security solution. Its algorithm adapts situationally to extract value from any blockchain-based software, identifying vulnerabilities without any prior knowledge of exploit methodologies(unknown unknowns).
This is achieved through a two-step black-box optimization process involving discrete optimization, continuous optimization, algorithmic parsing, efficient stateful computation techniques, and an innovative feedback mechanism to exhaustively identify value extraction opportunities.
"_“I am very excited to witness Mamori's mission to bring advanced ML-integrated fuzzing techniques to smart contract security. Their automated search and light specification approach form the perfect complement to formal verification! We are closer than ever to a future where smart contract developers get all the following at the push of a button: Mamori's tools help eliminate programming errors while learning the contract specification, semantics-based formal verification tools prove the program correct wrt the specification, then the correctness proof is ZK-ed and posted on-chain as verifiable evidence. Programmatically verify, not trust, security audits!”"
Mamori identifies both commonalities and unique aspects of value extractions, stemming from the nature of discrete search spaces. This approach has broader applications beyond security, including maximum extractable value (MEV) searching and intent solving.
The founder, Andy, spent his early career in academia focusing ML applications in Economics. He later joined the blockchain space specialized in AMM mechanism design and found security is the major roadblock of crypto mass adoption. He innovates on the methodologies leveraging interdisciplinary knowledge to create a Pathfinder for Blockchain Value Extraction. Sometimes for his leisure, he loves Jazz, dance and chess.
Mamori plans to expand its team to iterate and optimize its architectural framework. The team leverages insights from academia and Web2 extensively, and invites top talent and projects to collaborate and drive progress together. To stay updated with Mamori, follow them at x.com/mamori_xyz
About Mamori
Mamori is a ML-enabled pathfinder for blockchain value extraction, that can be applied to zero-day economic exploits, MEV search and intent solving, addressing the challenges of scalability, automation, exploit detection relevance, and efficacy in the Web3 value extraction landscape. We apply interdisciplinary technologies and protect against the ’unknown unknowns’ vulnerabilities. To achieve this, we leverage algorithmic parsing techniques to establish smart contract sequences, utilize reproducible stateful computation techniques, and incorporate innovative and customizable algorithmic feedback mechanisms.
