Hi there 👋 I’m Shephin Philip, an AI Engineer passionate about LLMs, Agentic AI, and enterprise-grade Generative AI solutions.
I specialize in building AI agents, RAG pipelines, and scalable backend integrations that solve complex, real-world problems. My journey blends hands-on engineering with a drive to innovate and bring ideas to production.
Enterprise AI Solutions (Apr 2025 – Present)
- Built RAG-powered chatbots with Hugging Face + FAISS; improved groundedness via rerankers.
- Integrated Azure OpenAI Services into enterprise pipelines with FastAPI + Streamlit.
- Designed evaluation harnesses (latency, groundedness, recall) to monitor deployed LLMs.
Driving AI Innovation (Oct 2023 – Mar 2025)
- Developed and fine-tuned LLMs (GPT, BERT, LLaMA) for NLP workflows.
- Orchestrated multi-agent systems using LangGraph & AutoGen for task automation.
- Deployed containerized services on Azure, AWS, and GCP with Docker/Kubernetes.
AI-Powered E-commerce (Feb 2024 – Mar 2024)
- Delivered an Agentic AI system with LangChain for product search & recommendations.
- Built a Flask backend that connected frontend queries to LLM agents in real time.
- 🥇 Top 10 Finalist – Mercor TextBase AI ChatBot Challenge (2023).
- 🚀 Production Impact – Deployed multiple LLM + RAG pipelines powering enterprise use cases.
- 🖐️ Social Innovation – Built a real-time Indian Sign Language recognition system using CNNs.
I believe in the transformative power of AI agents — moving beyond models to autonomous systems that solve real-world business problems.
My guiding principles:
- 🎯 Deliver scalable AI systems that make it to production.
- 🤝 Foster collaboration & mentorship within teams.
- 📈 Push the frontier of LLMs, RAG, and multi-agent orchestration.
"In AI, building agents that act, adapt, and scale is the next step toward a more intelligent future."
