Vijay Lakshminarayanan is the founder of the decentralized finance (DeFi) protocol VCRED and the consumer AI company RNDM. ^{[1] [2]} His career spans software engineering at Oracle, strategy consulting at firms including McKinsey & Company and PwC, and founding ventures in the Web3 and artificial intelligence sectors. ^[2]​

Education

Vijay Lakshminarayanan holds a postgraduate MBA degree with a focus on Finance and Strategy from the Indian Institute of Management, Calcutta, which he attended from 2011 to 2013. Prior to his business studies, he earned a Master of Science (MS) in Computer Science from Purdue University between 2006 and 2008. His undergraduate education was at the University of Madras, where he completed a Bachelor of Engineering (B.E.) in Computer Science from 2001 to 2005. ^[2]​

Career

Early Career and Consulting

Lakshminarayanan began his career as a Software Engineer at Covansys (now part of CSC) in Chennai, India, from July 2005 to June 2006. Following his master's degree in the United States, he joined Oracle in the San Francisco Bay Area as a Staff Engineer, a role he held from June 2008 to May 2011. After completing his MBA, he transitioned into management consulting, joining McKinsey & Company in September 2013. He started as an Associate in Chennai before moving to the San Francisco office, where he was promoted to Senior Engagement Manager, working at the firm until June 2018. He then served as Director of Digital Transformation at PwC in the San Francisco Bay Area from June 2018 to April 2021, where he focused on advising clients on technology strategy and implementation. ^[2]​

VCRED

In May 2021, Lakshminarayanan founded VCRED, a decentralized finance (DeFi) protocol, and served as its CEO. ^[2] The project was established to address capital inefficiency in the DeFi ecosystem, which largely relies on over-collateralized lending models. ^[3]​

Initial Concept and Launch

​VCRED was initially conceptualized and introduced as a protocol to bring flash loans to the Avalanche blockchain. ^[4] In a Medium post from early 2021, Lakshminarayanan described VCRED as the "first to bring flash-loans to Avalanche C-Chain," demonstrating a working prototype on the Fuji testnet. ^[4] A flash loan is a type of uncollateralized loan that is borrowed and repaid within a single blockchain transaction; if the repayment does not occur, the entire transaction is reverted, ensuring the safety of the lender's funds. ^[3] The stated goal was to enable capital-efficient DeFi strategies such as arbitrage, collateral swaps, and self-liquidations. ^[4]​

The following video, published by Lakshminarayanan, demonstrates the VCRED prototype on the Avalanche testnet.

<div> <div> <div> <iframe>

</iframe> </div><div> A 2021 demonstration video of the VCRED flash loan protocol prototype. </div> </div> </div>

Evolution and Technology

Over time, VCRED's vision expanded beyond flash loans. The project evolved into a protocol for uncollateralized and under-collateralized lending through a system of credit delegation. ^[2] This model was designed as a three-party marketplace consisting of lenders who deposit capital, borrowers who are whitelisted to access loans without full collateral, and "Vault Creators" who vet the borrowers and provide first-loss capital to underwrite the risk. ^[3]​

By early 2024, the project was described as an AI-powered liquidity layer. ^[5] This iteration of VCRED utilized automated, bot-driven vaults to execute sophisticated market-making strategies on decentralized exchanges. The protocol offered different vaults with distinct risk profiles, such as the "Atlas Vault" for delta-neutral market-making and the "Ganesh Vault" for more directional, high-reward strategies. ^[6] The stated purpose of this AI-driven approach was to democratize access to strategies typically reserved for quantitative hedge funds. ^[7]​

Lakshminarayanan chose the Avalanche blockchain for VCRED, citing its high throughput, low fees, fast transaction finality, and EVM compatibility as key factors for building the protocol. ^[3]​

Funding

While initial reports described VCRED as bootstrapped, later information from industry data platforms indicated that the project secured outside investment. VCRED is reported to have raised $1.5 million in a seed funding round on September 21, 2022, with Crypto3 Capital and GreenHorns Capital as lead investors. ^[6] Other sources list Avalanche's ecosystem fund, Blizzard, and AK44 Venture as investors in the company. ^[8]​

RNDM

As of late 2025, Lakshminarayanan's public profile indicates he is the Founder of a new venture named RNDM. This company's stated mission is to accelerate the adoption of consumer-facing AI agents. Associated projects under the RNDM umbrella include Atlas and Jenius. His self-described roles in this new context are "Trader" and "Backend Dev". ^[1]​

Interviews

AI, Liquidity and DeFi according to Vijay Lakshminarayanan #01

The interview DeFi, Liquidity and AI with Vijay Lakshminarayanan was released on the Blockchain Recorded Podcast on Feb 3, 2024. The episode presents Lakshminarayanan’s account of his work in decentralized finance, including his observations on liquidity conditions, automated trading systems, and the integration of artificial intelligence in blockchain based markets.

In the recording, he outlines his entry into the Web3 sector, beginning in 2019, after previous experience in software engineering for automotive and trading applications. He reports that his early involvement included participation in hackathons, where he developed prototypes that later contributed to the establishment of VCRED, a project built on the Avalanche blockchain.

Throughout the interview, Lakshminarayanan identifies recurring structural characteristics in decentralized finance, including liquidity fragmentation, the operational complexity of decentralized limit order books, and the limited availability of advanced trading strategies for general users. He refers to concepts such as arbitrage, market depth, and exchange architecture to outline how decentralized markets function and how these aspects influence pricing dynamics.

He describes VCRED as an AI based liquidity system designed for use on decentralized exchanges. According to his explanation, the system incorporates time series models and agent based decision processes oriented toward financial market data. These components operate together with a data storage structure and smart contract framework intended to automate selected trading and liquidity management tasks.

The interview also records his description of VCRED’s development path, beginning with an automated flash loan prototype and evolving into a broader liquidity focused mechanism. He outlines several challenges encountered during this process, including the construction of reliable AI models, communicating complex system behavior through simplified interfaces, and recruiting technically specialized contributors.

Toward the end of the episode, Lakshminarayanan presents his view of potential developments in the 2030s. He outlines a scenario in which autonomous software agents could perform financial operations across decentralized networks with continuous availability and emphasis on privacy. He states that such systems may work alongside human decision making, and that decentralization would remain a relevant organizing principle for these environments. ^[10]

Building Atlas on Aptos: Notes from an Interview with Vijay Lakshminarayanan #02

An interview published on the YouTube channel Aptos Developers on Nov 26, 2025, presents Vijay Lakshminarayanan describing his transition from work centered on Ethereum to the development of Atlas, a DeFi platform incorporating automated agents on Aptos. In the discussion, he outlines the technical considerations that influenced his decision to adopt the Move language. He identifies Move’s resource-oriented structure as a point of interest, particularly its model in which contract logic and associated data are tied directly to accounts, differing from the storage and deployment structure commonly used in Solidity.

Lakshminarayanan remarks that Move introduces a learning process distinct from languages with more flexible syntax. He refers to updates introduced in Move 2.0, noting that revisions to method invocation and documentation organization modified some aspects of early development workflow. Throughout the interview, he compares the storage and ownership framework of Move with the EVM’s global storage approach, emphasizing that these architectural differences shape patterns of contract design and execution.

Atlas, the project highlighted in the interview, is described as a system built to manage stablecoin operations through agent-based mechanisms. According to Lakshminarayanan, the platform uses resource accounts to isolate user-specific strategies rather than maintaining a consolidated pool. He indicates that this structure supports automated processes executed by agents without direct human intervention, which in turn distributes activity across individual accounts.

The conversation also addresses how developers familiar with Solidity may approach the Move environment. Lakshminarayanan notes that starting with smaller contract interactions and relying on TypeScript SDKs can simplify initial testing. In his account, introductory documentation and guided examples contributed to a straightforward onboarding process.

The interview concludes with comments on the Aptos ecosystem’s social dynamics. Lakshminarayanan mentions a shift toward more competitive interactions and suggests that in-person events and targeted hackathons may contribute to sustained collaboration. He characterizes community organization as an element that influences knowledge exchange within the ecosystem. ^[11]