Building production RAG systems and autonomous AI agents that actually ship.
Currently wiring LLMs into business workflows at Hudson Pharma Β· Karachi, PK
- ποΈ Production RAG @ Hudson Pharma β 5β8 yr organisational knowledge base with natural-language retrieval across Sales, Finance, HR, and Operations (Python Β· LangChain Β· pgvector Β· AWS Bedrock Β· Vertex AI)
- π€ Multi-department AI agent suite β autonomous agents for cross-functional decision support, routed into Slack (LangGraph Β· Lambda Β· Slack API)
- β‘ AI workflow automation β transitioned 25 % of daily org workflows to AI-driven automation via n8n, MCP servers, and custom toolchains
RAG systems ββββββββββββββββββββ 90 %
LLM agent design ββββββββββββββββββββ 85 %
Agentic orchestration ββββββββββββββββββββ 80 %
MLOps / prod AI ββββββββββββββββββββ 70 %
Full-stack (Next.js) ββββββββββββββββββββ 60 %
| Project | Stack | Status |
|---|---|---|
| Organisational Knowledge RAG | Python Β· LangChain Β· pgvector Β· Supabase Β· AWS Bedrock | π Proprietary |
| Multi-Dept AI Agent Suite | LangGraph Β· Lambda Β· Slack API Β· Python | π Proprietary |
Production work lives behind closed doors. Architecture and design patterns are fair game in interviews β feel free to reach out.
I'm open to roles involving LLM systems, agentic AI, RAG architecture, and AI integration engineering.
DM on LinkedIn or drop a mail at mkgilal@gmail.com
Last updated April 2026 Β· Karachi, Pakistan