Tijesunimi O.

Key Skills

Software

Other

Top Subject Matter

Agentic AI workflows and LLM task automation

Machine learning training and computer vision (identity verification)

Custom LLM development (training, fine-tuning, evaluation)

Top Data Types

Document

Audio

Image

Top Task Types

Prompt + Response Writing (SFT)

Data Collection

Transcription

Freelancer Overview

Hackathon Participant — Agentic AI (Meta AI Developer Academy). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Certificate, Meta AI Developer Academy (2025) and Certificate, Robotics & AI Nigeria (RAIN) (2025). AI-training focus includes data types such as Document, Computer Code, and Programming and labeling workflows including Prompt + Response Writing (SFT), Evaluation, and Rating.

Labeling Experience

Hackathon Participant — Agentic AI (Meta AI Developer Academy)

OtherDocumentPrompt + Response Writing (SFT)

Built and engineered agentic AI workflows intended for iterative task execution and evaluation using Llama and LangChain. Participated in creating an AI agent prototype by designing prompts, coordinating tool-augmented steps, and testing behavior under hackathon constraints. Focused on producing reliable outputs through prompt orchestration and system evaluation loops. • Developed agentic workflows with Llama and LangChain • Designed prompt-driven steps for automated task completion • Collaborated on system design and evaluation during the prototype build • Iterated on prototype behavior under tight deadlines

2025 - 2025

Custom LLM Development — Python, PyTorch, Transformers

OtherDocumentData Collection

Designed and trained a large language model pipeline involving data curation, tokenisation, and fine-tuning steps. Performed benchmarking by applying statistical analysis and evaluation metrics across NLP benchmarks to compare model performance. Optimized model inference for deployment constraints to ensure trained behaviors could be validated and used effectively. • Curated training data and performed tokenisation • Selected model architecture and executed fine-tuning • Benchmarked results using evaluation metrics across NLP tasks • Optimized inference speed and memory footprint for deployment

2024 - 2025

AI & ML Developer Student — Robotics & AI Nigeria (RAIN)

Other

Completed structured AI and ML training focused on learning modeling workflows, evaluation practices, and deployment-oriented understanding. Applied supervised and unsupervised learning concepts alongside evaluation and benchmarking to validate model performance. Built supporting computer vision modules for identity verification using Python-based tooling. • Learned supervised/unsupervised learning fundamentals • Practiced model evaluation and real-world deployment concepts • Implemented computer vision identity verification modules using DeepFace • Applied ML performance evaluation thinking to training outputs

2024 - 2025

AI Study Assistant — Multi-Format Data Pipeline (Streamlit, Deepgram, Groq)

OtherAudioTranscription

Built a multi-format AI study assistant pipeline that transformed video inputs into structured notes and study podcasts using speech-to-text plus LLM inference. Used deterministic processing steps to keep transformation outputs reproducible. Leveraged speech recognition and rapid LLM inference to convert spoken content into structured study material. • Converted video speech to text via STT • Used LLM inference to generate structured notes and podcast-style outputs • Designed deterministic, reproducible transformation pipeline • Applied low-latency architecture using Groq and Deepgram components

2024 - 2024

Education

Work History