Mateeh U.

Key Skills

Software

Micro1

Don't disclose

Top Subject Matter

Fund management workflows and LLM training data preparation

AI connector testing

synthetic data simulation

Top Data Types

Text

Document

Computer Code Programming

Top Task Types

Prompt + Response Writing (SFT)

Function Calling

Question Answering

RLHF

Fine-tuning

Evaluation/Rating

Computer Programming/Coding

Data Collection

Freelancer Overview

LLM Engineer, Turing—Remote (United States). Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose and Micro1. Education includes Bachelor of Science, FAST-NUCES (2023). AI-training focus includes data types such as Computer Code and Programming and labeling workflows including Prompt + Response Writing (SFT) and Function Calling.

Labeling Experience

AI Engineer, Micro1—Remote (United States)

Micro1Function Calling

Validated AI connector integrations for CRM and ticketing platforms to ensure reliable automated data exchange. Simulated real-world conditions by generating synthetic business data to improve test coverage. Automated connector testing via Python scripts integrated into CI/CD to streamline continuous quality checks. • Validated connector uptime to 99.9% • Improved test coverage by 30% • Automated connector testing in CI/CD • Saved 10+ hours per week through automation

2025 - 2026

LLM Engineer, Turing—Remote (United States)

Don't disclosePrompt + Response Writing (SFT)

Led benchmark-driven workflows for agentic AI systems to reduce hallucinations and improve task success in fund-management scenarios. Authored and reviewed multi-language code used for LLM training datasets. Conducted evaluation-oriented iteration of prompts, scenarios, and model behavior for higher reliability across downstream tasks. • Benchmarked agentic AI performance • Reduced hallucinations by 7% • Improved success rate across fund-management tasks • Enhanced scenarios for fund onboarding and audit-trail workflows

2025 - 2026

Education

Work History