Pfizer Externship via ExternEstablished career background, now reentering the workforce after a break and ready to contribute with focus and reliability.
A concise portfolio of past work and outcomes I can draw on as I return to professional life, emphasizing tangible tasks and steady delivery.
.jpg)
Pfizer · ✅ Verified by Extern · ⏱️ In progress
Prototype AI-powered document intelligence with Pfizer—using OCR, LLMs, and RAG to automate real enterprise PDF workflows and build a standout portfolio project.
Established career background, now reentering the workforce after a break and ready to contribute with focus and reliability.
I am gbonju Adesanya, an established professional reentering the workforce after a career break. I bring prior experience and a fresh focus on rebuilding momentum and contributing reliably to projects and teams.
Externships
Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship
In progress
Experience
SENIOR CLOUD ENGINEER
Sitecore · May 2021 - Present
INFRASTRUCTURE ENGINEER
CaseWare IDEA · Apr 2019 - Apr 2021
IT ENGINEER
Nova Networks · Aug 2018 - Apr 2019
Education
University of Ottawa, Canada
MEng, Electrical and Computer Engineering · Class of 2019
Federal University of Agriculture Abeokuta
BEng (Honours), Electrical and Electronics Engineering · Class of 2015
Skills
✅ Verified by Extern · ⏱️ In progress
Prototype AI-powered document intelligence with Pfizer—using OCR, LLMs, and RAG to automate real enterprise PDF workflows and build a standout portfolio project.
The externship built an AI-powered document intelligence prototype combining OCR, LLMs, and retrieval-augmented generation to process enterprise PDF workflows. Work included researching model behavior, designing ingestion and retrieval pipelines, and producing demonstrable extraction and question-answering artifacts.
I produced a technical explanation of LLM internals, describing tokenization, vector embeddings, and transformer attention as the mechanism producing generated outputs.
A compulsory task asked how LLMs work. I described that text was split into tokens, tokens were embedded as vectors, and a transformer model applied attention with the prompt to produce generated outputs.
