Ahammed Nibras ahammednibras8

Ahammed Nibras

Software engineer focused on AI product infrastructure, backend systems, and production reliability.

I work across LLM features, cloud/backend services, workers, databases, observability, and test-heavy product code.

AI / LLM systems: RAG, knowledge collections, document search, LLM integration layers, agent workflows, token usage tracking, cost analytics, streaming generation UX, and local LLM inference experiments.
Cloud / backend: TypeScript, Next.js, Bun, PostgreSQL, Drizzle, Redis/BullMQ, Celery, Docker, AWS EC2, API design, background jobs, and multi-tenant services.
Observability: OpenTelemetry, Grafana Cloud, frontend telemetry, worker monitoring, structured logs, and production debugging.
Security and data correctness: RBAC, tenant isolation, privacy/DSR flows, audit trails, PDF/content validation before AI processing, and database migration safety.
Testing: Vitest/Bun tests, Playwright e2e, route/static tests, regression tests, typecheck/lint/build gates, and CI-focused validation.

Private client work across AI product features, backend services, infrastructure, and reliability.

Technical areas I have worked on:

Built RAG and document-aware AI workflows.
Integrated LLM APIs, agent flows, token tracking, and cost analytics.
Worked on backend services using TypeScript, PostgreSQL, Redis-backed queues, and background workers.
Added observability with telemetry, logs, dashboards, and worker diagnostics.
Improved security and data correctness around RBAC, tenant isolation, audit trails, and privacy-safe deletion.
Shipped test-heavy changes with unit, integration, e2e, typecheck, lint, and build validation.

llama-infrence-server - local LLM inference server work around llama.cpp, RAM use, throughput, and concurrency behavior.
applemusicdiscord - Apple Music metadata to Discord Rich Presence, written in Go.
kvstore - a small Go key-value store project.
secure-mcp-db - experiments around MCP and database access patterns.

Last updated: 02-Jul-2026