Project Name
Tempora Labs
Project Website
https://temporalabs.com [Coming Soon]
Project Brief
Summary
Tempora Labs is a pioneering AI x Web3 startup building DR HIRO, the Agentic DeFi Portfolio Copilot for Institutional Funds.
Problem We’re Solving
DeFi is a highly fragmented, dynamic, and fast-moving space. Many chains, tokens, and exchange platforms are being launched or sunsetted at an accelerating pace. By 2030, DeFi institutional investors collectively hold approximately $500 billion in tokenized assets, but will lose roughly $3 billion annually due to the operational costs of navigating this complex DeFi space. These operational overheads, in turn, lead to poor investment decisions, such as portfolio misallocation and missed opportunities, further compounding the operational losses.
Our Solution
Our solution is an agentic DeFi portfolio copilot called DR HIRO. DR HIRO first captures client-defined investment intents in natural language, including the clients’ portfolio preferences, risk parameters, and rebalancing/withdrawal strategies. It then recommends portfolio plans to clients, and once approved, DR HIRO will autonomously execute the plans onchain and monitor them 24/7. We aim to significantly streamline the DeFi execution process for Web3 institutional investors, enabling them to focus on higher-level investment strategies. Having a portfolio copilot like DR HIRO can significantly reduce investors’ operational overhead at a fraction of the cost, just as coding copilots like Cursor and Claude Code reduce programming hassles for software engineers.
Key Features
DR HIRO’s key features include the following:
- Our state-of-the-art (SOTA) hierarchical multi-agent system design within DR HIRO that emulates a quant hedge fund, with specialized agents like data curators, research & risk analysts, portfolio manager, (onchain) execution trader, and a managing director (orchestrator).
- Our in-house agent framework, called nodeai, enables us to rapidly build and stack agents into a multi-agent framework such as DR HIRO. We are in the process of filing a provisional patent for nodeai.
- DR HIRO’s portfolio module that allows users to a) ask questions about b) edit/modify the portfolio c) deploy the portfolio live onchain d) track and monitor once deployed.
- An investable portfolio asset universe for DR HIRO to craft portfolio plans and execute them. So far, we have a list of yield-generating stablecoins and stablecoin vaults, but we will continue to expand the universe (including tokenized RWAs and native cryptos) to expand the range of possible DR HIRO portfolio combinations.
- DR HIRO natural language chat interface and dashboard UI.
Target Users & Market
Our target market is crypto institutional funds, including the following categories:
- Web3-native crypto funds, treasuries, family offices, neo-banks, and onchain allocators
- Incoming major TradFi funds seeking exposure to crypto opportunities but lacking the DeFi expertise to execute these strategies
DR HIRO can offer the following benefits to its institutional clients the following investment services:
- Recommend more opportunistic assets and portfolios in order to boost yields, especially for stable or delta-neutral positions.
- Automate portfolio deployment, execution, and monitoring onchain.
- Monitoring risk analytics and applying risk management capabilities (i.e., drawdowns), providing the optimal risk-adjusted returns rather than purely the maximal returns.
What makes your project unique
Our project is unique in the DeFAI or AgentFi in the following ways:
- Created a more custom or personalized portfolio tailored to individual clients’ intents, including risk, return, and liquidity preferences, rather than managing a wholesale pool of assets like in vaults.
- The broad range of potential assets and their allocation levels effectively enables the creation of an unlimited number of portfolios tailored to specific clients.
- The high level of interactivity and multi-step conversation between the DR HIRO agent and the client guides them through the entire portfolio lifecycle, from onboarding, planning, execution, rebalancing, and withdrawal.
Our team has an exceptional background across AI x Web3 x TradFi and can leverage our experiences, especially in AI and TradFi, to further differentiate DR HIRO from the other agent builders in the DeFi space.
How will you integrate 0G?
Technical implementation details + Which features will use 0G services
We plan to integrate the following from 0G:
1) 0G Storage Integration: First, we will migrate DR HIRO’s conversational history, logs, portfolio memory, and other metadata from our AWS S3 to the Log Layer for immutable storage. Second, we will store DR HIRO’s data lake, including market data, onchain metrics, sentiment, and risk analytics, into the Key-Value Layer so the DR HIRO agent can dynamically retrieve from this layer directly during inference rather than pulling it real-time data feed API. First, we will migrate DR HIRO’s conversational history, logs, portfolio memory, and other metadata from our AWS S3 to the Log Layer for immutable storage. Second, we will store DR HIRO’s data lake, including market data, onchain metrics, sentiment, and risk analytics, into the Key-Value Layer so the DR HIRO agent can dynamically retrieve from this layer directly during inference rather than pulling it real-time data feed API.
2) 0G Compute: We will utilize 0G’s global GPU and compute network for DR HIRO to scale its agentic inference calls amongst many users. Later, we will also develop our own custom, fine-tuned model for DR HIRO operations using the training compute services. For example, tool/function calling procedures.
3) 0G Product and Tools: We will leverage other 0G upcoming products, tools, and marketplace, as a way to further boost DR HIRO’s portfolio management activities and better interact with the 0G chain and infra stack.
Timeline for integration
Here is our estimated timeline for 0G integration:
- Week 1-2: Architectural proposal about how to migrate our tech stack over to 0G
- Week 3-4: Migrate DR HIRO chat logs and metadata from AWS S3 to the 0G Storage Log Layer
- Week 4-5: Migrate and build-out DR HIRO data lake to 0G Storage Key-Value Layer
- Week 6-8: Migrate foundational and fine-tuned models that DR HIRO uses for inference calls to 0G Compute
Note: It is possible implement several of these task items in parallel, which can further condense this timeline.
Expected benefits from using 0G
By being part of the 0G Guild and building on the 0G stack, we are able to greatly benefit from the following:
- Compute Scale: We have the compute capacity to run DR HIRO at scale, supporting more users and more sophisticated agentic usage (portfolio) flows.
- Lower Cost: We can achieve this compute capacity at 90% lower cost than centralized compute providers such as AWS (which we currently use).
- Security, Privacy, and Transparency: 0G storage is cryptographically committed, and every operation can be tracked and verified. This is especially important for us, as we handle conversational session data. Additionally, 0G compute provides TeeML for trusted execution without compromising private data, such as wallet keys.
- Marketability and Distribution: Being part of the 0G ecosystem can enhance our reputation and trustworthiness, as we have taken security, privacy, and transparency seriously. This is especially important for our GTM and distribution push towards institutional investment clients.
Additional Links
GitHub Repository: Tempora Labs · GitHub
Demo/Prototype: https://youtu.be/14absEd-MjI
Social Media:
-
Twitter/X: @TemporaLabs
-
Discord: kevrc
-
Telegram: @kevrc