Teska U.

Key Skills

Software

No software listed

Top Subject Matter

Customer support

content generation

and knowledge-grounded data extraction using LLMs

Top Data Types

Text

Top Task Types

Prompt + Response Writing (SFT)

Freelancer Overview

AI Prompt Engineer, DataMind Technologies. Brings 6+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include LangChain, Pinecone, and OpenAI API. Education includes Bachelor of Science, University of Nigeria, Nsukka (2020) and N/A, DeepLearning.AI (2023). AI-training focus includes data types such as Text and labeling workflows including Prompt + Response Writing (SFT).

Labeling Experience

AI Prompt Engineer, DataMind Technologies

TextPrompt + Response Writing (SFT)

Engineered and iterated 300+ production prompts for customer support, content generation, and data extraction workflows using LLMs. Built RAG-based prompt grounding to improve factuality and decrease hallucinations by linking outputs to proprietary knowledge bases. Created a structured prompt evaluation framework using automatic metrics to systematically validate prompt quality and reduce QA turnaround time. • Designed few-shot and instruction-based prompt variants • Implemented RAG pipeline with LangChain and Pinecone for knowledge grounding • Evaluated prompt effectiveness using BLEU, ROUGE, and G-Eval • Collaborated with ML engineers and PMs to translate requirements into prompt specifications

2023 - Present

Junior AI/NLP Specialist, BrightAI Solutions

TextPrompt + Response Writing (SFT)

Developed few-shot and chain-of-thought prompting strategies for summarization and classification tasks across multiple enterprise deployments. Conducted systematic A/B testing of prompt variants and documented results for team adoption. Fine-tuned open-source LLMs using LoRA to improve domain-specific performance metrics. • Authored prompt variants for summarization and classification • Ran A/B experiments to measure prompt impact • Documented findings in a shared knowledge base for broader use • Performed LoRA-based fine-tuning of LLaMA and Mistral models

2021 - 2022

Education

Work History