
AI marketing intelligence platform
Compass
A production platform for client intelligence, content generation, workflow automation, and human review using multi-provider LLMs.
- NestJS
- TypeScript
- PostgreSQL
- OpenAI
AI Engineer · Applied Researcher · Founder
I build evidence-grounded LLM applications, multi-agent systems, and retrieval platforms—and research how to make AI explainable, reliable, and safe to deploy.
Profile
Reliable, evidence-grounded AI—from published research to production.
Final-year Computer Science student, AI engineer, and published researcher working on trustworthy and explainable AI, multi-agent systems, Retrieval-Augmented Generation, medical AI, and production software engineering. I founded Konexium, an AI-native B2B research platform for African markets, and contribute to the Anserini and Pyserini information-retrieval toolkits.
My work spans backend architecture, model integration, retrieval and evaluation, human-in-the-loop workflows, and interpretability—built to be checked, not just trusted.
Now building · Early access
Founder · September 2026 — Present
Sell into Africa with evidence.
An AI-native B2B research platform for teams selling into African markets. Its autonomous agent, Baks, researches companies, verifies decision-makers, and links every fact back to the source it came from.
Selected roles
Konexiumkonexiumai.com ↗
Apping Technology Sdn. Bhd.Kuala Lumpur, Malaysia
Albukhary International UniversityMalaysia
Selected systems

AI marketing intelligence platform
A production platform for client intelligence, content generation, workflow automation, and human review using multi-provider LLMs.

Enterprise RAG assistant
An enterprise RAG assistant for organization-scoped document search, evidence retrieval, and configurable AI providers.

Conversational sales assistant
A production conversational sales assistant integrating backend services, AI workflows, and user-facing review.

Human-approved outreach automation
An attended LinkedIn outreach system that drafts fact-only messages, requires human approval, and verifies every recipient twice before sending.

Explainable clinical decision support
A HER2 breast-cancer decision-support system combining histopathology classification, Grad-CAM explanations, LLM recommendation drafts, and clinician approval.

Intelligent publishing automation
An AI-driven social publishing pipeline combining content generation, workflow orchestration, and browser/device automation.
Current direction
Trustworthy, explainable AI for high-stakes decisions and real-world workflows.
Explore researchResearch output

Zurinahni Zainol, Bakary Gibba, Altagi Abdallah Bakheit Abdelgadir, Almonzer Abdallah, Md Serajun Nabi, Basheer Riskhan
Smart, Inclusive, and Sustainable Digital Societies, CRC Press, pp. 162–174 · Co-author
Machine-learning analysis of behavioral and study factors associated with academic performance.

Bakary Gibba, Sami Salama Hussen Hajjaj
SITWE 2026 · Highlights in Web Engineering III, MMU Press · First author
LLM agents for extraction and explanation around a supervised matching model. Presented at SITWE 2026 on 30 September 2026.

Md Rashidul Islam, Bakary Gibba
arXiv:2603.23344 · Co-first author
Attention-enhanced segmentation with an explainability-oriented evaluation workflow.
Information retrieval
Completed all eight Castorini Foundations of Retrieval lessons — BM25, dense retrieval, learned sparse retrieval, retrieval evaluation, and manual score verification — and contributed reviewed documentation that was merged into both official repositories.
Capabilities
Academic foundation
Bachelor of Computer Science (Hons.)
Albukhary International University · Kedah, Malaysia
Expected November 2026 · CGPA 3.50 / 4.00
Verified credentials


