Develop
Backend systems
Designing product-critical APIs, workers, data flows, and operational services that stay boring under pressure.PythonGoFastAPIPostgreSQLRedis
About me
Iturnmessy😭workflowsintoclearsystems.APIs,queues,search,andagents🤖shouldstayboringunderpressure.Lessnoise.Moresignals⚡teamscanacton.
WhatI do
Automate
AI workflows
Turning scattered research, enrichment, search, and generation loops into reliable agent-facing systems.RAGLangGraphSearch APIsCeleryLLMs
Shape
Product engineering
Clarifying the system, the interface, and the operational contract so teams can ship with less guesswork.PlanningArchitectureObservabilityDX
My career and experiences
Senior Software Engineer
SubscriptionPro
NOW
Own a multi-channel AI concierge platform for the European luxury-travel market front to back: the agent loop, the retrieval and memory layers, the channel integrations, the operator console, and the AWS infrastructure under all of it.
Full Stack Software Engineer
SoftStandard Solutions
2025
Built an eight-agent sales system on one explicit state machine (company finder, lead finder, email finder, enrichment, verification, scoring, campaign, reply handling), with retry and backoff per stage and a single core exposed over both REST and MCP.
Software Engineer
Humana Apparels Ltd (Ha-Meem Group) — ready-made garments
2024
Built ProTrac, a real-time production-tracking system for one of Bangladesh's largest garment manufacturers, on Django REST and Next.js. Replaced day-late manual reporting with live bottleneck, DHU and plan-versus-actual signals, surfaced while a shift could still be corrected.
Software Engineer
Hours Media Ltd
2023
Built and ran backend services, automation APIs and background workers for client products (reselling, micro-investing, e-commerce) in Python and Next.js on PostgreSQL, working directly with clients to turn requirements into scope.
RAG
PostgreSQL
Go
Pinecone
Kubernetes
Pydantic
Vector DB
LLMs
FastAPI
LangGraph
Django
Systems for AI workloads
Tools I reach for when the system has to hold.
AI and Agents
AI/search
RAG, agents, source-aware retrieval.
Agentic WorkflowsMCP (Model Context Protocol)RAG
Service edges
APIs
Service edges and background workers.
PythonGo
System path
Build path
Search, APIs, data, product.
AI and AgentsService edgesReliability base
Reliability base
Data core
Storage, queues, deploys, and recovery paths.
SQL / PostgreSQLRedis (Streams)AWS (ECS Fargate, Secrets Manager, KMS)
Reliability base
Product surface
Interfaces where workflows become usable.
SQL / PostgreSQLRedis (Streams)
Working set
Practical stack
A compact slice grouped by layer.
AI and AgentsService edgesReliability baseReliability base
