Full-Stack Developer | AI Bots, LLM Apps & RAG | Next.js, Node.js, Python, I'm a full-stack developer specializing in AI bots and LLM-powered applications — not just the bot itself, but the complete product around it: frontend, backend, database, integrations, and deployment. Tech stack Frontend: React, Next.js (App Router, SSR/ISR), TypeScript, Tailwind CSS, shadcn/ui Backend: Node.js (Express, NestJS), Python (FastAPI, Django), REST and GraphQL APIs, WebSockets for real-time chat AI / LLM: OpenAI API (GPT-4o, function calling, structured outputs), Anthropic Claude API, prompt engineering, streaming responses, token/cost optimization RAG & data: embeddings, vector databases (Pinecone, pgvector, Chroma), chunking strategies, semantic search over PDFs, docs, and knowledge bases Bots & channels: Telegram Bot API, Slack (Bolt SDK, slash commands, event subscriptions), webhook architectures, WhatsApp via Twilio Databases: PostgreSQL (with Prisma), MongoDB, Redis for caching, queues, and rate limiting Automation: Python-based automation, n8n / [login to view URL] workflows, scheduled jobs, third-party API integrations (Stripe, CRMs, Google Workspace) Deployment & ops: Docker, VPS (Ubuntu, Nginx), Vercel, AWS (EC2, S3, Lambda), CI/CD with GitHub Actions, logging and error monitoring I focus on real systems that hold up in production — clear scope, fast execution, clean handoff. Not research demos or fragile prototypes. Most engagements start with a small pilot delivered in 1–2 weeks, then scale once the value is proven. What I build -AI & LLM bots (Telegram, Slack, internal assistants) with conversation memory, tool/function calling, and admin dashboards -Custom AI chatbots embedded in websites and SaaS products, with streaming UI and usage analytics -Document and PDF AI systems: ingestion pipelines, OCR when needed, vector search, cited answers (RAG) -Full web applications end to end — auth, payments, dashboards, APIs, deployment -AI copilots and internal tools that replace manual workflows (support triage, report generation, data entry) -Scraping and data pipelines feeding clean, structured data into your bot or database How I work -Clear scope agreed upfront, then fast execution -Business-outcome focused — every bot is measured in time saved or cost reduced -Founder-friendly communication: plain language, regular updates, honest timelines -Clean, maintainable, well-documented code your next developer can pick up easily -Start small, validate, then scale confidently Best fit clients -SaaS and tech startups adding AI features to their product -Founders who want a working AI bot, not a demo -Agencies adding AI capabilities for their clients -Ops-heavy businesses replacing manual workflows with AI assistants