I design, build, and deploy production-grade machine learning and AI systems that solve real business problems at scale.
My work spans classical ML, LLM-based systems, data pipelines, and MLOps, with a strong focus on robustness, performance, and real-world impact.

🎯 Currently focused on:

• LLM-powered applications (RAG, agents, evaluation frameworks)
• Scalable ML pipelines & feature engineering
• AI system design for data-intensive products

📍 United States | Open to Remote & Hybrid roles

• Applied Machine Learning: predictive modeling, NLP, feature engineering, experimentation
• Generative AI & LLMs: RAG pipelines, agentic systems, prompt engineering, evaluation
• Production ML Systems: model deployment, monitoring, CI/CD, data pipelines
• Data Engineering for ML: SQL-first design, ETL, analytics-ready datasets

I care deeply about clean abstractions, reproducibility, and measurable outcomes.

These repositories reflect how I think about real-world AI systems, not toy demos.

Tech: Python, FastAPI, OpenAI / OSS LLMs, Vector DB, Docker

• End-to-end Retrieval-Augmented Generation pipeline
• Document ingestion, chunking, embeddings, retrieval, and evaluation
• Production-ready API with monitoring hooks

🔗 Repo: llm-rag-platform

Tech: Python, SQL, Pandas, Scikit-learn, Airflow-style orchestration

• Data ingestion → feature engineering → model training → evaluation
• Emphasis on reproducibility and modular design
• Clear separation of data, training, and inference layers

🔗 Repo: ml-production-pipeline

Tech: Python, SQL, ML, Analytics

• Translating ambiguous business problems into ML solutions
• KPI-driven modeling and post-deployment analysis
• Strong emphasis on decision-making, not just accuracy

🔗 Repo: applied-ml-case-studies

Languages

• Python, SQL

ML & AI

• Scikit-learn, PyTorch (working knowledge)
• LLMs (OpenAI, open-source)
• RAG, embeddings, vector search

Data & Analytics

• Pandas, NumPy
• Feature engineering & data validation

MLOps & Infra

• Docker
• CI/CD concepts
• Model versioning & monitoring (foundational)

Cloud (working experience)

• AWS / GCP (ML workflows & deployment concepts)

• 🎓 Master’s in Data Analytics – Northeastern University
• 💼 Experience across consulting, product, and data-driven teams
• 🔍 Strong bridge between business context and technical execution

I believe strong AI engineers:

• Start with problem framing, not models
• Treat data pipelines as first-class systems
• Optimize for maintainability over cleverness
• Measure success using business and user impact, not just metrics

• 💼 LinkedIn: https://www.linkedin.com/in/pranshu-kumar
• 📧 Email: [email protected]

If you’re building data-intensive or AI-powered products, I’d love to collaborate or chat.

⭐ If you find something useful here, feel free to star a repo or start a discussion.