Sikirullah I.

AI & ML | Agentic systems | AI Security and Governance

Most AI projects don't fail because of bad code — they fail because nobody checked if the results can actually be trusted. I build and validate intelligent systems: agentic pipelines, LLM-powered workflows, and the evaluation frameworks that make them safe to act on. My work sits at the intersection most engineers avoid: where technical results meet real-world accountability. Recently, my focus has expanded into agentic AI engineering and AI security — designing multi-agent systems with built-in quality loops, hallucination detection, bias auditing, and compliance filters. I don't just build pipelines; I build pipelines that check themselves. I've worked across computer vision, NLP, OCR pipelines, and clinical/academic research — including cell classification, ASR data, and publication-ready statistical analysis. Where I add the most value: — You need an agentic system that produces outputs you can actually trust and act on — You have a model or dataset and need to know if it's reliable before it goes anywhere near a decision — You're writing a thesis, paper, or clinical report and need analysis that survives peer review — Your stakeholders need to understand what the AI actually found — and what it didn't — You need an honest assessment of where your AI system could fail, be gamed, or cause harm I'll tell you honestly if ML isn't the right solution for your problem. That's rarer than it sounds. Core skills: Agentic AI Engineering · LLM Systems & Evaluation · AI Security & Governance · Hallucination & Bias Detection · ML Validation & Evaluation · Healthcare & Academic Data Analysis · NLP · Computer Vision · Statistical Analysis · AI Risk Assessment · Research Reporting Tools & Technologies Agentic & LLM Systems: LangGraph · LangChain · Groq · OpenAI · Pydantic · FastAPI AI Security & Governance: Hallucination detection · Bias auditing · Compliance filtering · Adversarial input testing Validation & Evaluation: MLflow · Weights & Biases · Arize AI · SHAP · LIME · Fairlearn · Great Expectations ML & Deep Learning: Python · PyTorch · TensorFlow · Scikit-learn · Keras Data & Statistics: R · IBM SPSS · Pandas · NumPy · SciPy · Statsmodels NLP & Computer Vision: Hugging Face Transformers · NLTK · OpenCV Data Quality & Pipelines: Great Expectations · DVC · Pandas Profiling Visualization & Reporting: Power BI · Matplotlib · Seaborn · Plotly Cloud & Infrastructure: AWS · GCP · Google BigQuery · Azure