Saman A.

@samanazharr

Data Annotation and LLM QA Expert

Pakistan
English, Urdu
About me
I am a Machine Learning Engineer specializing in NLP, with hands-on experience building, evaluating, and deploying production-grade LLM systems. My background spans the full pipeline: from dataset curation and prompt engineering to security auditing, hallucination mitigation, and containerized deployment. Recently, my work focuses on LLM Evaluation & Alignment: refining model outputs through rigorous fact-checking, benchmark testing, prompt injection defense, and human-in-the-loop quality control.... Read more

Skills

s
samanazharr
Saman A.
Offline • 
Average response time: 1 hour

See my services

Data Labeling & Annotation
I will annotate your data for ai training
From$50 / projectMore details
AI Content Editing
I will humanize, fact check, and edit your ai generated content
From$30 / projectMore details

Portfolio

Work experience

Machine Learning Engineer

S • Self-employed

May 2025 - Present • 1 yr 3 mos

- Built and deployed LLM/NLP systems including RAG pipelines and CV–JD matching tools - Implemented end-to-end ML workflows: data ingestion, embedding, retrieval, inference, and evaluation - Deployed services using FastAPI, Docker, and Streamlit; tracked experiments with MLflow - Focused on production-oriented NLP use cases aligned with industry standards

Machine Learning Engineer

SlashNext • Full-time

Oct 2024 - Apr 2025 • 6 mos

- Applied NLP techniques to detect and prevent phishing attacks based on intent and content. - Developed models using TF-IDF vectorization and Random Forests, improving phishing detection accuracy in production. - Monitored and fine-tuned deployed models to optimize performance and reduce false positives. - Position impacted by post-acquisition restructuring.

NLP Engineer

Quarrio • Full-time

Jul 2022 - Sep 2024 • 2 yrs 2 mos

- Built NLP systems enabling natural language access to structured enterprise data (CRM, ERP, data warehouses) - Designed and maintained semantic parsing pipelines for accurate query understanding and intent resolution - Developed rule-based and hybrid NLP components to support high-precision, low-latency production use cases - Shipped new language features supporting complex analytical queries across business datasets - Analyzed usage and accuracy metrics to iteratively improve system performance - Created onboarding materials and trained 3 junior engineers, reducing ramp-up time by ~80%