Hi, my name is

Beauty Tasara.

I'm a Software Engineer

I’m passionate about researching large language models and finding ways to apply what I learn to real, everyday problems.

Say Hello

01.About Me

I’m a graduate researcher and engineer focused on robotics, computer vision, machine learning, and large language models, with a strong interest in turning technical ideas into practical, real-world applications. I enjoy working at the intersection of research and implementation, where mathematical thinking, intelligent systems, and thoughtful design come together to solve meaningful problems.

My recent work has included robotics and kinematics, image processing, NLP, and exploring how LLMs can be applied in everyday workflows. Whether I’m building in MATLAB, experimenting in Python, or studying new advances in AI, I’m motivated by creating systems that are both intelligent and useful

Outside of work and research, you can usually find me at the gym, planning meals, diving into technical papers, or working on ideas that connect AI with everyday life

Here are a few technologies I’ve been working with recently:

  • ▹Python
  • ▹Tensorflow
  • ▹MAtlab
  • ▹SLAM
  • ▹ROS TF
  • ▹Codex
  • ▹Cursor
  • ▹MCP
Profile Image

02.Where I've Worked

Software Engineer @ hail & cotton

August 2023 — December 2024

  • ▹Built an offline-first data collection system deployed to 500+ users, capturing text, image, and audio with low bandwidth sync, reducing manual data entry time by >90% and improving field operations scalability.
  • ▹Designed an automated PDF data-extraction pipeline in Python with ML models, eliminating manual processing and delivering structured datasets that powered data-driven decisions
  • ▹Developed and maintained features for ERP and inventory applications, supporting deployment, production troubleshooting, and server connectivity for business-critical workflows.
  • ▹Built an image classifier to identify tobacco leaf diseases, supporting consistent automated field inspections.

03.Some Things I've Built

Positional Bias Analysis with CoT & Prompt Repetition preview
P
Apr 14, 2026

Featured Project

Positional Bias Analysis with CoT & Prompt Repetition

Built NLP pipeline integrating OpenAI and OpenRouter LLM APIs with G-Eval model-based evaluation framework. Processed 100+ datasets with VADER sentiment analysis and cosine similarity metrics across multiple LLM providers (GPT-4o, Claude, Gemini). Estimating the effects of chain-of-thought prompting and prompt repetition on bias in multi-opinion summarization - evaluating LLM behavior systematically.
  • Javascript
  • ReactJS
  • NextJS
Tiny Search Engine preview
T
Jan 20, 2026

Featured Project

Tiny Search Engine

Developed a news aggregation system that fetches articles from RSS feeds (BBC, CNN), automatically generates AI powered summaries via OpenRouter API, and enables semantic search using TF-IDF vectorization with cosine similarity. Built with Flask REST API, JWT authentication, SQLite database. Background job scheduling, article indexing and user event tracking.
  • Javascript
  • ReactJS
  • Taiwind CSS
Autonomous Path Planning & Obstacle Avoidance preview
A
Apr 10, 2026

Featured Project

Autonomous Path Planning & Obstacle Avoidance

Engineered an autonomous differential-drive robot in MATLAB/Simulink using a hierarchical Stateflow FSM (11 states) integrating real time RGB color detection, odometry-based navigation, and reactive wall avoidance, achieving 100% task completion across all 6 color permutation trials (12/12 deposit points, 0 collisions) in a 5 m × 5 m simulated arena
  • MATLAB
  • Stateflow
  • Simulink
View the Archive

04.Get In Touch

Let's talk about everything and anything.

Say HelloGitHubLinkedIn