Tasmil Finance

Overview

While DeFi offers unmatched opportunities for yield generation and trading, it remains inaccessible for the majority of users. Complex strategies, high learning curves, and the need for constant monitoring make it difficult to capture returns without taking excessive risks. Yield farming, leverage trading, and cross-protocol strategies demand not only deep technical knowledge but also precise timing. Without intelligent automation, even experienced users struggle to consistently identify, execute, and manage profitable positions.

Vision

We envision a DeFi landscape where AI Agents handle the operational complexity - empowering every user to access sophisticated strategies, manage risks in real time, and maximize returns, regardless of their technical background.

Mission

Our mission is to merge AI-driven decision-making with the speed, scalability, and security of the blockchain, creating a platform that makes advanced DeFi operations effortless, transparent, and accessible to everyone.

About Tasmil Finance

Tasmil Finance functions as an intelligent DeFi co-pilot on the blockchain space, seamlessly connecting to a range of DeFi services. The platform offers two core modes of operation:

• Interactive DeFi Agent - A chat interface that lets users perform DeFi actions (trade, lend, stake, bridge, research) simply by chatting with the AI Agent.

• Automated Earning Agent - Ask, get recommendations, and choose bots or traders to copy trade or automate strategies, aiming for long-term passive income.

Together they let users act instantly or assign the AI to work continuously, turning user intentions into effective on-chain actions.

Key features

• Natural-Language Execution: Converts conversational input into secure DeFi actions across integrated protocols.

• Seamless Operations: Handles token swaps, lending, staking, bridging, and portfolio management without manual multi-step processes.

• AI-Driven Recommendations: Evaluates yields, risk factors and liquidity incentives to propose optimal strategies.

• Real-Time Insights: Delivers APY comparisons, market analytics, and governance summaries to support informed decision-making.

• Conversational Memory: Maintains context of previous actions to ensure smooth, continuous interaction.

• Automated Strategies: Provides access to algorithmic trading bots with configurable parameters for passive earning.

• Copy Trading: Enables mirroring of top-performing traders or AI agents with transparent performance and risk metrics.

• Secure Integration: Executes actions through audited smart contracts and API bridges, ensuring non-custodial control.

• Configurable Autonomy: Offers manual, semi-autonomous, and full autonomy modes with guardrails and auditability.

• Simulation and Testing: Runs backtests and simulations to validate strategies before live deployment.

• Scalable Infrastructure: Utilizes hybrid on-chain and off-chain architecture with credit-based resource allocation for responsiveness.

Integrated Functions

Interactive DeFi Agent

The Interactive DeFi Agent is an intelligent conversational interface that executes DeFi operations and provides real-time insights from user prompts. Acting as a personal financial assistant, it converts natural language commands into secure on-chain actions, simplifying complex tasks and making advanced DeFi more accessible.

The Interactive DeFi Agent delivers clear benefits for two key user groups:

• Retail DeFi Users

  • Simplifies complex actions into a single chat

  • Saves time on multi-step DeFi operations

  • Reduces errors through AI-assisted execution

• Newcomers to DeFi

  • Lowers entry barriers with an intuitive chat interface

  • Provides guided recommendations for key actions

  • Makes learning and using DeFi easier and less intimidating

Core Capabilities

a. Seamless DeFi Operations Executes token swaps, transfers, loans, and other actions across integrated protocols, automatically selecting optimal liquidity sources and confirming steps with the user. Example: “Swap 500 USDC to ETH at the best rate and then lend the ETH on a lending platform.”

b. Yield Farming & Strategy Execution Implements sophisticated yield strategies by evaluating interest rates, liquidity incentives, and risk factors to recommend and execute optimal approaches tailored to the user’s risk tolerance. Example: “Maximize yield on 1000 USDT with low risk,” prompting the Agent to suggest and deploy a stablecoin yield farm, liquidity pool, or delta-neutral position.

c. Research and Insights Provides data-driven analysis such as APY comparisons, portfolio allocation reviews, or governance outcome summaries to support informed decision-making. Example: “What’s the current APY for staking APT on the top platforms?” or “Analyze my portfolio’s asset allocation and suggest improvements.”

d. Cross-Chain Bridging Identifies secure and efficient routes for transferring assets between networks, leveraging supported bridge protocols to simplify multi-chain asset management. Example: “Bridge my USDC from Aptos to Ethereum,” prompting the Agent to select a suitable protocol and guide the transfer.

e. Personalized Support Maintains conversational memory to understand prior actions and contextual requests, ensuring intuitive, uninterrupted interactions. Example: After a swap, simply say “Now stake those tokens for me” and the Agent will use the swapped assets automatically.

Automated Earning Agent

While the on-demand chat agent is great for one-time actions, Tasmil Finance also enables users to put their investments on autopilot through pre-configured bots and copy-trading options. The Automated Earning Agent feature provides a conversational way to find and activate an automated strategy tailored to the user’s goals. In simpler terms, users can "hire" an AI trading bot or follow an expert trader by just chatting with the platform.

How it works: The user describes what kind of passive earning strategy they seek, and the AI assistant recommends either a set of algorithmic trading bots or skilled human traders to emulate. The user can ask questions about each option and, once satisfied, deploy the chosen strategy with a few clicks.

There are two primary categories of automated strategies offered:

Algorithmic Trading Bots

These are AI-driven or algorithm-driven bots that execute a particular trading or investment strategy 24/7 on behalf of the user. Tasmil offers a variety of bot strategies to cater to different risk appetites and market conditions. For example:

• Grid Trading Bot: Executes a “grid” strategy on spot or futures markets, buying and selling within preset price intervals to profit from volatility.

• Dollar-Cost Averaging (DCA) Bot: Gradually accumulates a target asset by purchasing small amounts at regular intervals, smoothing out the entry price over time - ideal for users who want to build a position without worrying about market timing.

• Funding Rate Arbitrage Bot: In perpetual futures markets, this bot captures passive yields by long-short strategies that exploit funding rate differences. Essentially, it will long an asset on the spot market and short the perpetual futures when the funding rate is positive (or vice versa), earning the funding payments with minimal net price exposure.

• Portfolio Rebalancing Bot: Automatically maintains a desired asset allocation. For instance, if a user wants to keep a 50/30/20 split between three tokens, the bot will periodically rebalance by trading, ensuring the portfolio stays close to the target allocation and taking profits on assets that grew above their share.

• Each trading bot offers configurable parameters with default settings. Through the chat interface, users receive tailored recommendations, then the platform automatically deploys and connects the chosen bot to begin operation.

Copy Trading & Social Trading

For users who prefer to rely on human expertise, Tasmil’s system can recommend top-performing traders whose strategies can be automatically mirrored. The AI assistant maintains a database of performance metrics for various seasoned traders or community-created AI agents (from Tasmil’s own marketplace of strategies).

When the user expresses interest in copy trading, they might say, "I want to follow a trader who consistently earns passive income on blue-chip tokens." The AI could reply with a short list of traders or strategy profiles, including statistics like:

• Total PnL (Profit and Loss): e.g., Trader X has made +50% returns over the last 6 months.

• ROI and Win Rate: e.g., Trader Y averages 8% ROI per month with a 70% trade win rate.

• Risk Metrics: e.g., Max drawdown of Trader X’s portfolio is 15%, Sharpe ratio is 2.0 indicating good risk-adjusted returns.

Users can request performance-based recommendations, and upon selecting a trader to copy, Tasmil’s infrastructure links their account to mirror the trader’s actions through smart contracts or API integration.

Under the hood

System Architecture

Underlying Tasmil’s user-friendly interface is a sophisticated AI and blockchain integration. This section outlines how Tasmil works behind the scenes to deliver the above experiences securely and efficiently.

AI Agent Engine

At the core of Tasmil Finance is an advanced AI engine behind a user-friendly interface. It converts natural-language input into executable DeFi operations. The engine combines large language models with financial models, processing real-time on-chain data and selected off-chain signals to deliver context-aware recommendations and actions.

Smart Contract and API Integration

The platform integrates directly with Aptos-based DEXs, lending protocols, staking modules and bridges through secure, audited smart contracts. When off-chain data or centralized exchange access is required, Tasmil employs oracles and API bridges to extend its operational reach while maintaining trust and security.

Wallet Control and Security

The architecture is non-custodial. Transactions must be signed by connected wallets, ensuring asset control remains external to the AI agent. The system supports three operational modes with built-in guardrails:

• Manual Mode: The agent only generates recommendations and requires explicit approval for execution.

• Semi-Autonomous Mode: The agent executes routine, low-risk tasks automatically while requiring approval for major or high-risk actions.

• Full Autonomy Mode: The agent executes strategies within predefined boundaries without requesting approval for each action, subject to strict rules and limits configured in advance.

Comprehensive logs and real-time notifications ensure full auditability of the agent’s actions.

Simulation and Testing

Strategies and bots can be backtested or simulated using historical data before live deployment. This allows evaluation of expected performance, volatility and potential risks, increasing confidence in execution outcomes.

Scalability and Infrastructure

Tasmil Finance employs a hybrid architecture. Critical operations run on-chain for transparency, while computationally intensive AI tasks are processed off-chain on secure servers. A credit-based system allocates compute resources to ensure fairness and responsiveness across all operations.

Agentic Frameworks

Intent–Reasoning Synthesis

Tasmil Finance begins by interpreting the user’s intent from natural language input and synthesizing a strategic plan. The platform’s AI analyzes the user’s goals, matches them against a library of available decentralized finance (DeFi) agent strategies, and determines the optimal execution route to achieve the desired outcome. To enrich its reasoning, the system employs retrieval-augmented generation (RAG), drawing on up-to-date on-chain data (such as blockchain state, market prices, and protocol metrics) and off-chain information. This combined data context provides the agent with strategic insight, allowing it to formulate a well-informed plan aligned with the user’s objectives.

Isolated Validation

Before any real capital is executed, the proposed strategy undergoes isolated validation through an integrated large language model (LLM). In this stage, the LLM simulates the agent’s reasoning and potential outcomes in a safe, sandboxed environment, allowing the system to identify any logical errors or unfavorable results ahead of time. The simulation’s results - including expected actions and projected outcomes - are then presented to the user for approval.

If the user chooses not to proceed, the framework archives the entire decision pathway (including the reasoning behind the strategy and the user’s response). This archived data serves as a learning example for the system, enabling it to refine future strategy recommendations based on past rejections or adjustments.

OrEx Agent (Orchestration + Execution)

Upon user approval, the framework transitions to the OrEx Agent stage, where the validated strategy is translated into a sequence of concrete on-chain actions and orchestrated for execution on the blockchain. The OrEx agent interacts directly with DeFi protocols to implement each step of the plan – for example, executing token swaps, providing liquidity, or initiating lending operations as required.

All execution is handled autonomously by the agent, which encodes and dispatches transactions with the correct sequence and parameters.

Retrain Memory

In the final stage, all user decisions, strategy pathways, and transaction outcomes are logged into a dedicated memory layer, forming a historical record that is continuously used to retrain and fine-tune the platform’s AI models. By learning from actual outcomes and user feedback (both approvals and rejections), the system incrementally improves its future performance.

The memory-driven feedback loop ensures that Tasmil Finance’s agentic framework is not static, but rather evolves with each interaction - enhancing strategic insight, safety, and alignment with user goals as the system gains experience.