Leakhena Suon

• Outlier is a platform that connects subject matter experts to help build the world's most advanced Generative AI.

• Concrete Tasks:

• Evaluate the performance of two LLM AI models by comparing their responses and providing constructive feedback on their strengths, weaknesses, and areas for improvement related to AI's general and programming knowledge.

• Everyday Activities

• Work as an AI trainer consultant on projects for various Outlier's clients, focusing on both general and specific programming knowledge, including C++, and Python.

• Generate the prompt and review the model's response to ensure the code is trustworthy and meets the prompt's requirements.

• Correct the code in the response if it is unhelpful or does not follow the prompt's instructions.

• For the code refactoring category, provide a new solution that outperforms the AI solution.

• Analyze each model's response in terms of both general and programming knowledge by assessing key criteria, including instruction following, conciseness, truthfulness, localization, writing style, and tone. The goal is to identify the better performing model by scoring each response based on these criteria and refining the weaker model's response to improve its quality.

• Achievements

• Consistently delivered high-quality results on client projects, earning a promotion as a high-quality contributor.

• This project aims to help asset managers automate the process of portfolio analysis and predict the optimal portfolio.

• Concrete Tasks:

• Developed a website to showcase data visualization related to portfolio analysis and optimal portfolio predicting by simply providing the input including the asset name, start date, end date, and total number of assets to invest in.

• Everyday Activities:

• Developed and tested this automation application by displaying the data such as close price history, covariance between equities, correlation between equities, expected return of each asset, and portfolio optimization showing the best assets to be chosen from the available options to maximize the return and minimize the risk.

• Achievements:

• Successfully merged quantum finance technology with Django framework to display data visualization of portfolio optimization on the front page.

• Access to this application: https://soma.biz/

• Technology and Tools:

• Quantum finance using Qiskit library , Visual Studio Code, Docker

• Django framework , DevOps practices: GitLab CI/CD pipeline

• Etops provides services regarding digitalization in the financial industry, data collection, and analytics.

• Concrete Tasks:

• Developed investment proposal for ESG Portfolio Optimization

• Everyday Activities:

• Wrote a master's thesis using the requirement engineering and software architecture method from high to low-level software design, on the topic of Investment Proposal of ESG Portfolio Optimization

• Developed investment proposal with iteration development process

• Achievements:

• Completed master's thesis and obtained master's degree in computer science and Software Engineering

• Completed the first demo version of ESG Portfolio Optimization

• Technology and Tools:

• React, Node.js, PostgreSQL

• Visual Studio, MySQL Workbench, pgAdmin, UML

• Microservice architecture , Domain-Driven Design