Hi there, I'm Prashant 👋

Built an intelligent email assistant that saves time by turning simple subject lines into actionable summaries. Using a custom LangChain tool and secure Gmail integration, the agent autonomously searches through thread history to find the right conversation. It then processes the content to deliver a clear, concise digest of the discussion, highlighting key decision points and necessary action items so you don't have to read every message.

Built an end-to-end Expense Classifier that demonstrates the full lifecycle of custom LLM development. I curated a specialized dataset to fine-tune a Google Gemini model on Vertex AI, tailoring it to recognize specific spending patterns better than generic models. The system is deployed on a scalable endpoint and features a simple Streamlit interface, allowing users to input data and get instant classifications from the custom model.

A Gen AI app that can generate image based on prompt & analyze images, video & audio.
Tools used:
Python
Vertex AI APIs
StreamLit
Google's LLM models like gemini & imagen

Built an application that turns any blog post into an interactive knowledge base. It scrapes and processes the webpage into a FAISS vector store, allowing users to ask questions that are answered accurately using GPT-4o and the specific blog context. The entire pipeline is fully instrumented with LangSmith, providing real-time tracing and visibility into every step of the retrieval process.

Deployed a containerized Python chatbot on Kubernetes to demonstrate seamless application lifecycle management. The project established a reliable update workflow: containerizing new versions, pushing them to Docker Hub, and executing rolling updates via K8s manifests. This setup ensured that the upgrade from v2 to v3 went live instantly without disrupting the service.

Developed a custom chatbot POC that bridges generative AI with rigorous MLOps practices. Using OpenAI's Assistant APIs and Python, I built the core application and automated its deployment pipeline using GitHub Actions. The project is containerized with Docker and orchestrated on a Google Kubernetes Engine (GKE) cluster, utilizing Kustomize and Kubernetes Secrets for secure, scalable management.

Demo video of a Gen AI app developed using,
1. Ollama - setup locally
2. Gemma 2b LLM model
3. StreamLit
4. Langchain
5. Langsmith

This video demonstrates a text translator Gen AI app developed using the following tools:
1. GROQ library & gemma 2 LLM model
2. StreamLit
3. Langchain
4. Python
The app uses LLM model to translate the given text & explain the meaning of each of the translated word.

Developed a Corrective RAG system using LangGraph that actively quality-checks its own data before answering. The application retrieves documents from a local FAISS database and evaluates their relevance to the user's query. If the local data is insufficient, the system automatically transforms the query and performs a web search to ensure the final response is accurate and grounded.