AI Engineer | GenAI & RAG Systems | Full Stack Dev (React/Node/MongoDB) | Multi-Agent LLM Pipelines | Building Production-Grade AI Apps | CS @ JIIT ’28
Bachelor of Engineering, Computer Science
jaypee institute of information technology
Join kartik and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal BossAI Engineer | GenAI & RAG Systems | Full Stack Dev (React/Node/MongoDB) | Multi-Agent LLM Pipelines | Building Production-Grade AI Apps | CS @ JIIT ’28
Most students learn AI by watching tutorials. I learn it by shipping systems that break, fixing them, and shipping again. I'm a Computer Science undergrad at JIIT who builds production-style AI applications — not toy chatbots. My work sits at the intersection of GenAI architecture and full-stack engineering: multi-agent pipelines that research and write reports on their own, RAG systems that turn hours of meeting audio into searchable knowledge, and automation agents that triage business workflows without a human in the loop. A few things I've built: 🔹 RecapAI — A 6-stage audio intelligence pipeline (ingestion → transcription → summarization → extraction → RAG indexing → chat) that converts meetings, calls, and YouTube links into queryable insights. Dual-engine transcription (Whisper + Sarvam AI) supports bilingual Hindi-English speech. Fault-tolerant, chunk-level recovery ensures one bad audio segment never kills a full run. 🔹 Autonomous Multi-Agent Research System — A 6-agent architecture (Splitter → Search → Reader → Planner → Writer → Critic) that decomposes a research question, retrieves and cleans web sources in parallel via Tavily, and self-critiques its own output before finalizing a cited report — closing the feedback loop without human review. 🔹 AI Expense Tracker — Full-stack (React/Node/MongoDB) app with JWT auth, receipt-image parsing, and Gemini-powered personalized budgeting insights generated directly from transaction history. 🔹 AI Feedback Automation Agent — n8n + Gemini pipeline that classifies incoming customer feedback and auto-routes it across Slack/email/Airtable — zero manual triage. I've also completed JPMorgan Chase's Software Engineering Job Simulation (Kafka, Spring Boot, Spring Data JPA) and solved 150+ DSA problems on LeetCode. Currently deepening my skills in production ML systems and agentic AI architecture, and actively looking for AI/Software Engineering internships where I can build things that matter from day one. 📩 Open to internships, collaborations, and interesting technical conversations — feel free to connect or DM.
Noida
Bachelor of Engineering, Computer Science, jaypee institute of information technology