Trashu Vashisth
Building Production-Grade AI Agents & RAG Systems
New to Contra
Trashu is ready for their next project!
The Problem: Sales teams waste 60% of their time researching leads instead of closing them.
The Solution: I built a custom Agentic AI Pipeline that automates deep-dive business intelligence and lead scoring.
Key Technical Highlights:
Multi-Agent Architecture: Built using CrewAI, featuring a 'Business Intelligence Specialist' (for real-time research) and a 'Senior Sales Director' (for strategic scoring).
High-Speed Intelligence: Powered by Llama 3.3-70B for near-instant reasoning and decision-making.
Real-time Web Scoping: Integrated Tavily AI to fetch live revenue data, employee counts, and market positioning.
Enterprise Storage: A robust SQLite backend to manage lead pipelines with a sleek Streamlit dashboard.
Smart Throttling: Engineered custom rate-limiting and token-trimming logic to ensure 99.9% uptime even under heavy API constraints.
How it works:
Simply enter a company name and URL. The AI agents scour the web, analyze the company's "AI potential," calculate a priority score (0-100), and even write a personalized sales pitch—all in under 30 seconds.
0
7
I built a professional, end-to-end AI Receptionist system designed to automate clinic appointment management. This isn't just a chatbot; it's an AI Agent that can reason, use tools, and manage a live database autonomously.
Key Contributions:
Agentic Reasoning: Integrated CrewAI with Llama 3.3 (Groq) to enable the agent to understand complex user intents (Booking vs. Cancellation) and relative time (e.g., "next Tuesday at 3pm").
Autonomous Tool Use: Developed custom Python tools that allow the agent to verify real-time availability in a SQLite database and execute atomic transactions without human intervention.
High-Performance Backend: Built a robust API using FastAPI to handle asynchronous requests between the AI agent and the database.
Premium Dashboard: Designed a modern, Glassmorphic UI using Tailwind CSS that provides a real-time sync of the clinic’s schedule.
The Result:
A seamless, hands-free system that reduces administrative overhead by 100%, allowing clinic staff to focus on patients while the AI handles the entire scheduling lifecycle.
Tech Stack:
Python, CrewAI, Groq API, FastAPI, SQLite, Tailwind CSS
0
25
An autonomous AI system that turns a simple voice command into a deep-dive research report in seconds. No typing, no manual searching.
Key Highlights:
Voice Control: Uses Speech-to-Text for hands-free research triggers.
Multi-Agent Intelligence: Powered by CrewAI & Llama 3.3 (Groq) to find, verify, and summarize live web data.
Voice Synthesis: Delivers an instant audio summary via ElevenLabs.
Automated Export: Generates a professional PDF report automatically.
Tech Stack: CrewAI, Groq, ElevenLabs, Streamlit, DuckDuckGo API.
1
43
Developed a highly responsive AI Voice Agent using Vapi that handles real-time conversations with exceptional clarity. The agent is designed to engage users naturally, gather specific information during the call, and accurately extract that data for further use. The voice quality for both the user and the bot is seamless, making the interaction feel professional and human-like
2
91
Developed a production-grade Retrieval-Augmented Generation (RAG) system specifically designed to automate the analysis of complex Environmental, Social, and Governance (ESG) reports. This tool bridges the gap between static LLMs and the dynamic, data-heavy requirements of legal and sustainability compliance. [1 (https://www.youtube.com/watch?v=wkYPcMtwlN8)]
Key Features & Capabilities
Intelligent Document Processing: Automatically handles large, unstructured PDF/Word ESG reports, extracting critical clauses and metrics in seconds.
Fact-Grounded Q&A: Uses a RAG architecture to ensure all answers are strictly based on the uploaded documents, virtually eliminating AI hallucinations.
Compliance Mapping: Cross-references internal company data with global frameworks like CSRD, GRI, and TCFD to identify gaps or inconsistencies.
Audit-Ready Traceability: Every insight generated includes direct citations and excerpts from the source files, providing a clear "paper trail" for legal teams.
Automated Drafting: Capability to draft legal summaries, notices, or internal policy updates based on analyzed ESG risks
Note: The 'Slaughter and May' branding in the sidebar is for UI/UX demonstration purposes only, showcasing how the tool integrates into a top-tier law firm's environment.
#AI #RAG #LegalTech #ESG #Python #LangChain
1
139
Developed a full-stack RAG-based E- Commerce AI chatbot using React.js and Tailwind CSS that suggests the perfect laptop from a live catalog. Integrated ChromaDB with BGE Embedding models to provide highly accurate, context-aware product recommendations and instant technical support."
Key Highlights:
Smart Laptop Recommendations: Uses Semantic Search to match user needs (gaming, coding, etc.) with real-time specs.
Advanced Tech Stack: Powered by LangChain for orchestration and BGE models for superior data retrieval.
Modern UI/UX: Built a responsive, clean interface using React.js and Tailwind CSS.
Zero Hallucination: Ensures all suggestions are strictly grounded in the available product inventory.
2
1
226
Built a highly scalable Retrieval-Augmented Generation (RAG) chatbot designed to interact with private datasets/PDFs. Unlike standard LLMs, this system minimizes hallucinations by retrieving real-time context from a local knowledge base before generating responses.
Key Features:
Semantic Search: Implemented Vector Embeddings to perform high-speed similarity searches across thousands of document chunks.
Smart Retrieval: Integrated a retrieval pipeline using LangChain to fetch the most relevant context for user queries.
Source Citation: Configured the bot to provide source references from documents, ensuring data transparency and accuracy.
Optimized Performance: Used FAISS/Chromadb for efficient vector storage and retrieval.
4
292
Developed a high-precision Resume Parser using a custom-trained RoBERTa model, specifically fine-tuned for Named Entity Recognition (NER) tasks. This tool automates the extraction of critical information from unstructured resumes with deep learning accuracy.
Key Features:
NER-Based Extraction: Accurately identifies entities like Name, Experience, Skills, Education, and Contact Info.
RoBERTa Architecture: Leverages Transformer-based embeddings for superior contextual understanding compared to traditional parsers.
JSON Output: Seamlessly converts complex resume layouts into structured JSON format for easy database integration and ATS (Applicant Tracking System) workflows.
High Accuracy: Trained to handle diverse formatting and professional jargon.
3
253
Developed an automated pipeline to handle bulk recruitment workflows. This AI-powered tool processes multiple PDF resumes simultaneously, using advanced parsing to extract candidate details with high accuracy. It intelligently structures unstructured resume data into a clean, downloadable CSV format, capturing essential metrics like skills, experience, and contact info without manual data entry.
4
315
Built an intelligent RAG-based chatbot designed to simplify complex financial analysis. In the project demo, the AI deep-dives into Apple’s annual reports, extracting key fiscal metrics and providing real-time insights through natural language queries. It transforms dense financial filings into actionable data using advanced document retrieval
2
3
297