Dahiru Ibrahim Dahiru
👋🏾Hi, I'm Dahiru. An AI Engineer. I get a thrill from solving real-world problems with f = Wx + b and that's why I'm porting my understanding of AI to the web. Read my Blog See my GitHub
Experience
Location
Tech stack
A powerful 2D to 3D pipeline. Uses a dynamic fusion of YOLOv8, GroundingDINO, Segment-Anything2 & Gaussian Splatting to transform 30-second 360-degree videos into interactable 3D models.
Story-Story is an engaging storytelling app designed for children. It takes the rich histories of Nigerian and African figures and turns them into child-friendly, factual stories with vibrant illustrations and captivating narratives
Get your daily summaries of Nigerian news in style. A news aggregator that scrapes news from PunchNG's RSS Feed, summarizes them, categorizes them, and distributes to users accordingly via email.
TLDR the T&C uses an LLM to understand the contents of Terms and Conditions documents and provides a user-friendly chatbot interface for users to ask questions and receive answers.
Simple toolkit for converting Open Images Dataset v4 (OIDv4) annotations to YOLO format.
Mood Scope detects emotions of people in an image by embedding the image into a vector space using CLIP and comparing which emotion vector it is closest to.
Fire Detection System Using Gas and Temperature Sensors.
Send e-mail alerts when a symbol's quote reaches a set threshold. Uses Deriv API
Enhanced AI capabilities by reducing hallucination in Video-LLaMA, developing multi-agentic workflows with LangGraph, creating a MongoDB retrieval tool, and fine-tuning Gorilla-LLM for enterprise API retrieval.
Involved optimizing the company's flagship AI models for real-world efficiency to achieve a multiple times inference speedup with TensorRT
Focus on advancing detailed object-level reconstruction by using Instant NGP, a Neural Radiance Fields platform from NVIDIA and integrating it with YOLOv8 coupled with SAM to identify individual objects within a scene
Involved collaborating with researchers from Colgate University to create a facial recognition system that allows for the non-intrusive analysis of harbor seals and other marine mammals, enabling the tracking of their population and movements.
Graduated 4.20/5.00. Relevant coursework: Control engineering, digital electronics, machine learning, embedded systems, digital communication, computer networking
Best way to reach me is through: suhayrid6@gmail.com