Team
Socrates
Project Concept
No description has been added yet.
Entry
Status: Submitted
Last saved: May 09 at 5:56 PM IST
Team Roster
Message board not available for this team yet.
Mounesh Kodangal Team Lead RSVP Approved
Student at Student
Mounesh designed the agent logic and orchestration layer. He built the OpenAI function‑calling pipeline that constructs a structured dashboard schema (a JSON array of widgets with type, title, data query, and position) from the user’s natural language prompt. He integrated CopilotKit’s backend runtime (CopilotRuntime) to stream the schema to the frontend as generative UI instructions. Mounesh also set up the data fetching layer that maps generic metric identifiers to actual data endpoints (simulated financial data), ensuring the agent can request information without hardcoding any datasource.
I’m an AI & ML undergraduate based in Hyderabad, focused on building practical Generative AI systems. I work on multimodal LLM applications, Retrieval-Augmented Generation pipelines, and AI-powered full-stack products. I enjoy combining machine learning with modern web technologies to create scalable, production-ready solutions. I’ve participated in national hackathons and contributed to large-scale AI projects, continuously pushing toward building impactful AI systems.
Interested in advanced AI agents, scalable LLM infrastructure, multimodal reasoning systems, and AI product architecture. Looking to collaborate on impactful AI startups, research-driven GenAI projects, and production-grade enterprise AI systems.
Currently building multimodal and RAG-based GenAI systems using Gemini, LangChain, and FAISS. Exploring advanced prompt engineering, Chain-of-Verification pipelines, and real-time AI agents. Experimenting with scalable LLM applications using Next.js and TypeScript, and improving evaluation frameworks to reduce hallucinations and increase response faithfulness.
Mittapalli Praneeth RSVP Approved
Praneeth owned the generative UI rendering engine and the frontend component architecture. He integrated CopilotKit’s useCopilotAction and useCopilotReadable hooks to allow the agent to dynamically mount dashboard components (charts, tables, metric cards) based on the LLM output schema. He built the reactive layout grid that adapts to any component set and implemented the interactive filtering/sorting logic using Recharts and shadcn/ui. Praneeth also handled the real-time data binding layer that connects agent-decided metric keys to live API responses.
Aspiring software engineer passionate about problem solving, DSA, and backend development. I enjoy learning core computer science concepts, building projects, and collaborating with developers who are focused on growth and innovation. Currently preparing for software engineering interviews while improving my skills in Java, algorithms, system design fundamentals, and development workflows.
Interested in backend engineering, scalable systems, interview preparation strategies, developer communities, startup culture, and collaborative learning. Looking to connect with engineers, mentors, and peers working on impactful products, open-source projects, or innovative startups.
Currently building DSA-focused Java projects and improving problem-solving skills for coding interviews and online assessments. Exploring backend development, APIs, and scalable system design concepts while practicing real-world problem solving through competitive programming and project-based learning.
Satwik Reddy RSVP Approved
Student at MLRIT
Set up the data fetching layer that maps generic metric identifiers to actual data endpoints (simulated financial data), ensuring the agent can request information without hardcoding any datasource.
I am a final-year B.Tech student in Artificial Intelligence and Machine Learning based in Hyderabad, India. As a backend-focused developer, I specialize in Java and Spring Boot, emphasizing high-performance systems, secure APIs, and database query optimization. Recently, I have been expanding my skill set by building AI agents and integrating them into traditional backend architectures. I also have hands-on experience deploying cloud-native applications on Google Cloud Platform using Cloud Run and Compute Engine VMs. I enjoy tackling complex algorithms, having solved over 600 DSA problems, and recently placed Top 20 at the GDG Agentathon for building a multi-agent system.
I am highly interested in the intersection of traditional backend engineering and Artificial Intelligence. Specifically, I want to learn more about developing and deploying scalable multi-agent systems, integrating LLMs with production-grade Java and Spring Boot backends, and exploring distributed systems and real-time data processing. I am eager to connect with professionals working on AI-driven backend applications, scalable cloud architecture, or intelligent fintech solutions.
Currently developing MAFA (Multi-Agent Financial Assistant) across three repositories (Spring Boot backend, FastAPI agents, and frontend). It is designed for automated trading and market research using LangChain and LangGraph. I am focusing on high-throughput agent queries, Redis caching to improve latency, and deploying the microservices on Google Cloud Platform to bridge the gap between AI agents and robust, production-ready backends.