n
naveedahmad963

Naveed A

@naveedahmad963

Custom AI, ML, Data Science systems for your business

Pakistan
Anglais
Certaines informations sont présentées en anglais.
À propos de moi
I build production-grade AI, machine learning, and data science solutions for businesses that need their tech to actually work — not just demo well. Services include: custom RAG chatbots, LLM fine-tuning (Llama 3, Mistral, Qwen), recommendation systems for e-commerce, predictive ML models, Python notebook debugging, and end-to-end data science pipelines. Stack: Python, PyTorch, TensorFlow, LangChain, FastAPI, Pandas, Scikit-learn, Hugging Face, Pinecone, Docker. Deployment-ready code, evaluation reports, and full ownership of what we build. Serious projects only. Message me first to scope.... Plus d’infos

Compétences

n
naveedahmad963
Naveed A
hors ligne • 
Temps de réponse moyen de 1 heure

Voir mes services

Développement de chatbots d'IA
I will build a custom rag chatbot and fine tune llm on your data
De295 $US / projetPlus de détails

Portfolio

Expérience professionnelle

Content_Lab UK

Machine Learning Engineer

Content Lab UK • Temps plein

May 2024 - Present • 2 yrs 1 mo

Lead ML engineer on the customer retention and revenue forecasting workstreams for a Series B B2B SaaS company in the customer success space. Own the full lifecycle from problem scoping with Customer Success and Finance leaders through model deployment and monitoring. Key work: ▪️Built and deployed a churn prediction model on usage telemetry, billing events, and support ticket sentiment, feeding daily risk scores into the CRM. Reduced voluntary churn by 19% over two quarters. ▪️Developed a customer lifetime value model that became the primary input to the Finance team's expansion revenue forecasts. ▪️Built an NLP layer on inbound support tickets to surface product issues at the category level, replacing a manual weekly review process. ▪️Stack: Python, scikit-learn, PyTorch, Hugging Face Transformers, FastAPI, AWS SageMaker, Snowflake, dbt, Airflow, MLflow.