Thanks to visit codestin.com
Credit goes to github.com

Skip to content
View intikhab49's full-sized avatar
🎯
Open for Work
🎯
Open for Work
21 followers · 19 following

Achievements

Achievement: QuickdrawAchievement: Pull SharkAchievement: YOLOAchievement: Pair Extraordinaire

Achievements

Achievement: QuickdrawAchievement: Pull SharkAchievement: YOLOAchievement: Pair Extraordinaire

Block or report intikhab49

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
intikhab49/README.md
Intikhab Azam, AI engineer and systems architect: voice AI agents, RAG chatbots, machine learning and automation. Open to contract work.

Intikhab Azam · AI Engineer & Systems Architect for Voice AI Agents, RAG, Machine Learning & Automation

I'm an AI engineer and systems architect. I design, build, fix and optimize AI agents, voice AI, RAG systems, machine-learning models and business automations that run in production: the kind that answer real phone calls, move real money and get used by people who don't care how they work.

role:      AI engineer · systems architect · automation engineer · backend developer
building:  voice AI agents (Vapi + Twilio) · RAG chatbots · LLM tool-calling agents
architect: multi-tenant SaaS · BFF auth · field-level encryption · event-driven pipelines
ai stack:  Claude · Claude Code · MCP · n8n · OpenAI · Gemini · OpenRouter · ElevenLabs · Deepgram
also:      ML validation + forecasting · computer vision · lead generation · data pipelines
obsessed:  LLM cost optimization: prompt caching, model routing, token budgets
shipping:  production systems: voice agents, CRMs, lead-gen engines, ML research tools

What I build

What I build: Vapi and Twilio voice AI agents wired into CRMs and payments; RAG chatbots with tool calling and memory; machine learning, face recognition and honest backtesting; lead generation, scraping and scheduled automation pipelines; multi-tenant system architecture; Claude, MCP and n8n agent workflows.

System architecture

Reference architecture for production AI systems: web, phone and Slack clients; Cloudflare Workers edge and BFF auth; NestJS and FastAPI services with job queues and n8n; LLM router with Claude, GPT-4o, small models and RAG; multi-tenant Postgres with row-level security, encrypted fields, object storage and Prometheus.

I design systems end to end, from the data model to the deploy pipeline. Patterns I use in production:

  • Multi-tenant SaaS: shared schema with tenant keys and Postgres row-level security
  • BFF authentication: httpOnly cookie sessions with token refresh owned in one place
  • Field-level encryption for sensitive healthcare data, with key rotation and idempotent backfills
  • Event-driven pipelines: webhooks, job queues, retries and idempotency keys
  • Edge compute: Cloudflare Workers for payment links, SMS and API glue between voice agents and Stripe
  • LLM routing: small models first, frontier models only when needed; cut one client's LLM spend by about 70%
  • Delivery: Dockerized services, Kubernetes + NGINX, Prometheus monitoring and CI/CD gated on end-to-end tests

Featured projects

EdgeProof: crypto trading strategy backtesting and ML validation with triple-barrier labels, purged cross-validation and deflated Sharpe ratio Vapi Voice Agent CRM: call-ops backend for AI voice agents with predictive dialer, auto-redial and payment-verified conversions Local Business Lead Finder: self-hosted lead generation engine past the Google Places 60-result cap with OpenStreetMap fallback SecureFace: face recognition with liveness detection and anti-spoofing using a head-pose challenge and rPPG pulse detection AI Wealth Advisor: RAG and tool-calling LLM agent for portfolio risk, VaR, Sharpe ratio and drawdown CryptoAion: BiLSTM with attention crypto price prediction API with FastAPI and WebSockets NPI Healthcare Lead Pipeline: NPI Registry to practice websites to e-mails, scheduled on GitHub Actions Kubernetes Flask microservices with NGINX, PostgreSQL, a logging service and Prometheus monitoring

All repositories, with details
Repository What it is Stack
edgeproof-crypto-trading-backtest Crypto trading backtesting & ML validation: triple-barrier labels, purged CV, deflated Sharpe and a planted-edge control Python · LightGBM · PyTorch
vapi-voice-agent-crm CRM & call-ops backend for Vapi AI voice agents: webhooks, predictive dialer, auto-redial, Stripe-verified conversions Node.js · Express · Twilio · Stripe
local-business-lead-finder Lead generation engine: grid search past the Google Places cap, OpenStreetMap fallback, e-mail enrichment Python · FastAPI · Streamlit
secureface-liveness-detection Face recognition access control with head-pose liveness, rPPG pulse detection and ChromaDB matching InsightFace · OpenCV · Streamlit
ai-wealth-advisor RAG + tool-calling financial assistant: VaR, Sharpe, drawdown, diversification scoring LangChain · Gemini · Flask
cryptoaion-price-prediction BiLSTM + attention crypto price prediction API with WebSocket streaming and JWT auth PyTorch · FastAPI
npi-healthcare-lead-pipeline NPI Registry → practice websites → e-mails, resumable and scheduled on GitHub Actions Python · GitHub Actions
kubernetes-flask-microservices Flask microservices on Kubernetes with NGINX, PostgreSQL, a logging service and Prometheus Flask · Kubernetes · Docker
kubernetes-flask-crud Flask + PostgreSQL behind NGINX on Minikube with NetworkPolicies Flask · Kubernetes · NGINX
flask-postgres-crud-dashboard User management CRUD app with a Chart.js analytics dashboard Flask · PostgreSQL · Docker
react-express-drizzle-crud Full-stack TypeScript CRUD with Passport session auth React · Express · Drizzle · Neon
devops-agency-website DevOps agency site with a pricing calculator and contact API React · TypeScript · Tailwind
crypto-price-prediction-api v1 of CryptoAion with BiLSTM and XGBoost research notebooks FastAPI · PyTorch
ubuntu-dev-tools-installer Interactive Bash installer for Python, VS Code, Docker, Jenkins, kubectl Bash
libncurses5-libaio1-ubuntu-deb Legacy libaio1 / libncurses5 / libtinfo5 .deb packages for newer Ubuntu Ubuntu · dpkg
cpp-queue-data-structure Array-based FIFO queue in C++ C++

Tech stack

Tech stack: Python, PyTorch, scikit-learn, Claude, Anthropic, MCP, OpenRouter, ElevenLabs, Deepgram, Retell AI, NotebookLM, LightGBM, XGBoost, NestJS, Next.js, Prisma, Drizzle, Neon, better-auth, MinIO, Cloudflare Workers, pnpm, n8n, Slack, Notion, Resend, Playwright, DrissionPage, Jest, NumPy, pandas, Jupyter, LangChain, OpenAI, Gemini, Hugging Face, OpenCV, Twilio, Vapi, FastAPI, Flask, Node.js, Express, Stripe, PostgreSQL, MySQL, SQLite, MongoDB, TypeScript, JavaScript, React, Vite, Tailwind CSS, Astro, Streamlit, C++, Docker, Kubernetes, NGINX, Prometheus, GitHub Actions, Linux, Bash, Cloudflare, Vercel

Live GitHub stats

Live GitHub stats: public repositories, contributions in the last 12 months, streaks, weekly contribution skyline and primary languages

Drawn daily by this repo's own GitHub Action. No third-party stats service.

Work with me

I take on contract and freelance work:

  • System architecture: multi-tenant SaaS design, auth and data security, scaling and cost reviews
  • AI agent workflows: Claude, MCP and n8n agents across Slack, Notion and CRMs
  • AI voice agents: Vapi / Twilio receptionists, outbound callers, call-to-CRM pipelines
  • RAG chatbots & LLM agents: knowledge bases, tool calling, memory, evaluation
  • LLM cost optimization: cutting token spend with caching, routing and smaller models
  • ML validation: finding out whether a model's backtest is real before money rides on it
  • Automation: lead generation, scraping + enrichment, CRM and payment integrations

Open to contract work: voice agents, RAG chatbots, LLM cost optimization, ML validation and automation

AI engineer · systems architect · software architect · solutions architect · multi-tenant SaaS architecture · Claude AI developer · MCP · n8n automation expert · voice AI agent developer · Vapi developer · Twilio voice bot · RAG chatbot developer · LLM agents · LangChain developer · machine learning engineer · algorithmic trading backtesting · computer vision · Python developer · FastAPI · automation engineer · lead generation automation · LLM cost optimization · freelance AI engineer

Popular repositories Loading

  1. crypto-price-prediction-api crypto-price-prediction-api Public

    v1 of CryptoAion: FastAPI service forecasting Bitcoin, Ethereum & altcoin prices with a PyTorch BiLSTM + attention model, CoinGecko and Yahoo Finance data, plus BiLSTM and XGBoost research notebooks.

    Jupyter Notebook 2 1

  2. kubernetes-flask-microservices kubernetes-flask-microservices Public

    Flask microservices on Kubernetes (Minikube + Flannel) and Docker Swarm: NGINX reverse proxy, PostgreSQL, logging service, Prometheus monitoring and NetworkPolicies — a hands-on DevOps project.

    Python 1

  3. cpp-queue-data-structure cpp-queue-data-structure Public

    Array-based FIFO queue in C++ with enqueue, dequeue, size and display — a data structures coursework exercise.

    C++

  4. GET-REPO GET-REPO Public

  5. ubuntu-dev-tools-installer ubuntu-dev-tools-installer Public

    Interactive Bash script to install or update Python, VS Code, Docker, Jenkins and kubectl on Ubuntu — detects installed versions and asks before changing anything.

    Shell

  6. flask-postgres-crud-dashboard flask-postgres-crud-dashboard Public

    Flask + PostgreSQL user management CRUD app with a Bootstrap dark UI, Chart.js dashboard and REST API — runs with Docker Compose.

    Python