Currently an Extern @Pfizer
Master's student with hands-on SQL and database architecture experience from a Breaking Games externship.
Portfolio items show SQL, schema design, and database architecture work completed during my Breaking Games externship and academic projects.
Breaking Games · Aug 2026 · ✅ Verified by Extern
Turn six messy CSVs into an analytics database and a Q4 dashboard that drives real business decisions. SQL + Claude.
.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.
Master's student with hands-on SQL and database architecture experience from a Breaking Games externship.
I am Abubakar Ibrahim, a graduate computer science student pursuing a master's degree. I have some practical work experience and completed an SQL and database architecture externship with Breaking Games. I am building technical skills as I transition into a formal career.
Externships
Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship
Pfizer
Experience
Volunteer Digital Skills Trainer
IMH MAGs Virtual Skills Acquisition Program · August 2026 26ndash; Present
Senior IT Instructor & Project Coordinator
Megabrains Infotech Institute, Gombe · 2019 26ndash; Present
Field Researcher & Project Coordinator
Deliver Nigeria Project / Munir Dankwambo Foundation
Education
University of the People
Master of Science (M.Sc.) in Computer Science (In Progress)
Gombe State University
Bachelor of Science (B.Sc.) in Computer Science (Second Class Upper Division)
Skills
✅ Verified by Extern
Turn six messy CSVs into an analytics database and a Q4 dashboard that drives real business decisions. SQL + Claude.
The externship converted six messy CSV files into an analytics database and a Q4 dashboard. Work began with a survey of the six data sources, then moved to schema design, ETL, and SQL queries to prepare analytics tables. Deliverables included a cleaned relational schema and query set for the dashboard.
I produced an inventory of six raw data sources, built SQL queries against the provided SQLite file, and delivered answers to five business questions including top products by units and revenue and total Meta ad spend.
✅ 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 prototyped AI-powered document intelligence for enterprise PDF workflows using OCR, LLMs, and RAG techniques. The work surveyed core ML, deep learning, generative AI and NLP concepts, investigated computer vision and OCR methods for text extraction, and examined pharmaceutical documentation practices relevant to automated processing.
I documented LLM internals and training methods, explaining tokenization, vectorization, Transformer self-attention, next-token training, and fine-tuning approaches including RLHF.
I examined LLM internals, explaining tokenization, vector embeddings, the Transformer self-attention mechanism, next-token training on large corpora, and fine-tuning with methods like RLHF to improve safety and alignment.
The project required cleaning structured and text data, handling JSON, and preprocessing scanned documents. I wrote Pandas routines to fix inconsistent fields, used text libraries to normalize content, and applied OpenCV filters (denoising, CLAHE, adaptive thresholding) to produce OCR-ready images.
The project parsed a 3‑page SDF with PyMuPDF, extracted word-level text and bounding boxes, grouped split multi-word labels by coordinate proximity, and used regex anchors to locate manufacturing, effective, revision, and expiration dates for structured output.