I build AI agents, backend systems, and enterprise automation that solve real engineering problems.
My current focus is on Agentic AI, multi-agent architectures, tool-using LLMs, MCP integrations, and production-grade backend systems built with Python and Django.
Building AI systems that can reason, use tools, interact with real software, and keep humans in control of consequential actions.
Multi-agent workflows, repository-aware coding agents, tool calling, human-in-the-loop systems, and MCP-powered integrations.
Scalable APIs, financial workflows, authentication, optimization systems, background processing, and production application architecture.
LangChain, LangGraph, RAG, structured tool usage, LLM orchestration, Gemini, Groq, and local models through Ollama.
Custom business workflows, ERP integrations, accounting processes, automation, deployments, and enterprise application development.
A repository-aware multi-agent software engineering system that can inspect codebases, implement scoped changes, run validation, and safely prepare GitHub pull requests.
User Request
│
▼
Developer Agent
│
├───────────────┐
│ │
▼ ▼
Code Agent GitHub Agent
│ │
▼ ▼
Repository Human Approval
Inspection │
│ ▼
▼ Branch → Commit → Push → PR
Code Changes
│
▼
Tests / Lint / Build
Highlights
- Multi-agent task routing
- Repository-aware code inspection
- Constrained file editing
- Automated test/lint/build execution
- Human approval before Git operations
- GitHub MCP integration
- Gemini, Groq, and Ollama support
Tech
Python LangChain LangGraph MCP Gemini Groq Ollama GitHub
A voice-enabled Agentic AI food ordering system where users interact naturally while AI manages the ordering workflow.
The browser handles speech interaction while the Django backend controls business rules, cart state, checkout, AI orchestration, and persistence.
Highlights
- Voice-first AI interaction
- Tool-using ordering agent
- Server-controlled cart and checkout rules
- Persistent order state
- Production-oriented Django backend
- Separate React frontend
Tech
Django LangChain Gemini React PostgreSQL REST API
Frontend: food-order-agent-frontend
💰 Ravani
A privacy-aware personal finance platform designed around monthly budgeting, household expenses, savings, reconciliation, reporting, and financial planning.
Highlights
- Household budgeting
- Savings goals
- Planned vs actual spending
- Bank reconciliation
- Financial reports
- Role-based household access
- Google authentication
- Automated reminders
- JSON/CSV exports
- Production deployment architecture
Tech
Django Django REST Framework React TypeScript PostgreSQL Neon Docker
I'm particularly interested in engineering problems around:
- 🤖 Multi-Agent Architectures
- 🧠 Agentic AI
- 🔗 Model Context Protocol
- 📚 Retrieval-Augmented Generation
- 🛠️ AI Developer Tools
- 🧑💻 Repository-Aware Coding Agents
- ⚡ Backend Architecture
- 🏢 Enterprise Automation
- 💳 FinTech Systems
- 📊 ERP / Business Applications
LangChain · LangGraph · RAG · MCP · Tool Calling · Multi-Agent Systems
Python · Django · Django REST Framework · FastAPI · Celery · Redis
React · TypeScript · JavaScript
Frappe Framework · ERPNext · Financial Workflows · Business Automation
PostgreSQL · MySQL · MariaDB · MongoDB · SQLite · Redis
Docker · Git · GitHub · Linux · Render · Neon · Frappe Cloud
I'm currently going deeper into:
- Multi-agent software engineering systems
- Autonomous coding workflows
- Repository-aware AI agents
- MCP-based tool integrations
- Production RAG architectures
- Agent memory and context management
- Human-in-the-loop AI systems
- AI automation for enterprise workflows
I prefer building AI systems where:
LLMs handle reasoning.
Tools handle deterministic operations.
Business rules remain application code.
Humans retain control over high-impact actions.
That separation makes AI systems safer, easier to test, and more reliable in production.
I'm interested in opportunities around:
Agentic AI · AI Engineering · Python Backend · Django · Multi-Agent Systems · Enterprise AI
📧 Email: sallahudinawan8@gmail.com
