Hilary

HILARY AZIMOH
APPLIED AI ENGINEER | EXPLAINABLE LLM SYSTEMS • RAG • RESPONSIBLE AI

📞 [open contact info](look above in the "contact info" section) ✉️ [open contact info](look above in the "contact info" section) A Amsterdam,
Netherlands •
Remote-friendly
🔗 https://hillariaa.github.io/ai-portfolio/
https://hillariaa.github.io/grc-portfolio-site/

PROFILE

Applied AI Engineer focused on building safe, explainable, and CORE SKILLS
reliable LLM-based systems for enterprise and regulated AI & LLM Engineering:
environments. Hands-on experience designing Retrieval-Augmented ●​ Retrieval-Augmented Generation
Generation (RAG) pipelines, grounding models in verified data, and (RAG)
enforcing safe-failure behavior to reduce hallucinations. Strong ●​ LLM grounding & hallucination
Python skills with a background in AI governance, cybersecurity, and mitigation
responsible AI system design. ●​ Prompt engineering for constrained
and explainable outputs
●​ Embeddings & semantic search
EXPERIENCE ●​ Source-attributed and auditable AI
responses
Applied AI Engineer / Founder — HC Core Tech ●​ Safe-failure design (“I don’t know”
behavior)
Netherlands / Remote
2025 – Present Machine Learning & NLP:
●​ Design and build safe, explainable AI systems for policy, ●​ Transformer-based models
compliance, and internal knowledge use cases. ●​ Sentence embeddings
●​ Translate ambiguous requirements into structured, testable ●​ Text similarity & retrieval
AI system logic with predictable behavior. ●​ Model evaluation for reliability and risk
●​ Apply security-by-design and Responsible AI principles
across system architecture and implementation. Programming & Tooling:
●​ Python
Applied AI & LLM Engineering ●​ Hugging Face Transformers
●​ Designed and implemented an Explainable ●​ FAISS
Retrieval-Augmented Generation (RAG) system to answer ●​ Git & GitHub
policy questions using verified internal documents. ●​ API-based system design
●​ Built semantic retrieval pipelines using sentence embeddings
and vector search (FAISS) to ground LLM outputs strictly in Cloud & Enterprise Foundations:
source material. ●​ Azure fundamentals
●​ Implemented safe-failure logic, enforcing explicit “I don’t ●​ CI/CD concepts
know” responses when retrieved evidence is insufficient. ●​ Secure system design
●​ Designed prompts and system constraints to reduce ●​ Enterprise AI considerations
hallucinations and prevent confident but unsupported
answers.
 ●​ Attached source references to generated responses to
improve explainability, auditability, and trust.
●​ Evaluated trade-offs between model capability and reliability, Responsible AI & Governance:
selecting instruction-tuned models for predictable enterprise
●​ AI risk assessment & mitigation
behavior.
●​ Explainability and transparency
●​ Implemented the system end-to-end in Python, covering
principles
document ingestion, chunking, embedding generation,
●​ EU AI Act awareness
retrieval, and response generation.
●​ NIST AI Risk Management
●​ Version-controlled and documented the solution using Git
Framework
and GitHub to ensure reproducibility and technical review
readiness.
EARLIER CAREER SUMMARY

Security, Governance & Enterprise Context Prior experience spanning cybersecurity,
digital risk, compliance operations, and
●​ Designed AI governance and risk-assessment approaches technical coordination across regulated and
aligned with NIST AI RMF and EU AI Act principles. client-facing environments. Worked closely
●​ Supported compliance readiness activities including with engineers, product teams, and
documentation, access governance, and risk mapping. stakeholders to translate risk, security, and
●​ Applied minimal-disclosure principles to reduce information policy requirements into practical technical
controls, documentation, and system
leakage in AI-assisted systems.
processes. This background informs a
strong engineering focus on reliability,
explainability, and risk-aware system
EXPLAINABLE RAG SYSTEM FOR POLICY QA design.
Personal Project
EDUCATION
●​ Built an explainable RAG system that answers policy questions BSc (Hons) Computer Science (in
progress) — Open Institute of
strictly from verified source documents.
Technology (Remote)
●​ Implemented semantic retrieval using embeddings and FAISS to Sep 2024 – Dec 2026
ensure factual grounding.
Relevant coursework: Artificial
●​ Enforced safe-failure behavior when relevant context is
Intelligence, Machine Learning, Data
unavailable.
Science, Information Security
●​ Added source attribution to all generated answers to improve
transparency and auditability.
CERTIFICATIONS
●​ Designed prompts and constraints to reduce hallucinations and
overconfident responses. ●​ Google Cybersecurity Professional
●​ Implemented the full pipeline in Python, from document ingestion Certificate
●​ AWS Academy Graduate — Cloud
to response generation. Foundations
●​ IBM AI Engineering
(See portfolios linked above.) ●​ Responsive Web Design (SheCodes /
freeCodeCamp)

LANGUAGES
English • Russian • Dutch