PORTFOLIO / 2026BACKEND · SYSTEMS · FULL-STACKSTATIC / INTERACTIVE / ZERO TRACKERS

ENGINEERING-LED PORTFOLIO

I build the parts of a product that have to keep working.

I'm Faizan Saiyed — a full-stack engineer focused on backend APIs, realtime systems, asynchronous workflows, data, and the interfaces that make those systems useful.

Explore selected work↘

02SELECTED SYSTEMS

04SOURCE REPOS

01INTERACTIVE GUIDE

FS / 01LIVE PORTRAIT

FASTAPI

REALTIME

POSTGRES

FAIZAN SAIYEDFULL-STACK ENGINEER

ENGINEERING SIGNAL ONLINE

APISTATEJOBSSTREAMSVERIFICATION

00 / ORIENTATION

The framework is rarely the interesting part. The interesting part is deciding where state lives, what happens when work takes too long, what the client is allowed to trust, and how you prove the system behaves.

FS↗

01 / SELECTED WORKENGINEERING CASEBOOK

Projects presented as systems, not screenshots.

01 / ASYNC MEDIA SYSTEM SOURCE AVAILABLE

FrameFlux

Keep heavy media work out of the request path.

Uploads and media transformations become explicit job lifecycles with durable state, validation, progress, retries, and cancellation.

FastAPIPostgreSQLRedis / ARQFFmpeg

Backend repo ↗Frontend repo ↗

ARCHITECTURE / ASYNCFLOW 01

BROWSER

FASTAPI

STATEPOSTGRES

UPLOAD

VALIDATE

REDIS / ARQ

WORKER

FFMPEG

RESUMABLE STATEQUEUE + PROGRESSPROCESSING

THE INTERESTING PART

A resumable upload is a state machine. A media transform is a job lifecycle. The architecture makes both visible.

02 / REALTIME OBSERVABILITY SOURCE AVAILABLE

Telemetry

Give live system state a model the interface can trust.

Synthetic telemetry flows through one backend-owned model for anomaly detection, persistence, alert lifecycle, authenticated streaming, and a live dashboard.

FastAPIWebSocketsPostgreSQLReact / TypeScript

Backend repo ↗Frontend repo ↗

RUNTIME / REALTIMEFLOW 02

SIGNAL

TELEMETRY MANAGER

STATEANOMALY + ALERTS

DETECT

PERSIST

POSTGRES

WEBSOCKET

DASHBOARD

SYNTHETIC INPUTBOUNDED PERSISTENCELIVE STREAM

THE INTERESTING PART

The frontend is not the authority. The backend owns alert lifecycle, roles, persistence, simulation state, and the live stream.

02 / SYSTEMS THINKINGTHE REPEATED MOVES

The way I approach engineering is visible in the projects above.

Separate expensive work.

Long-running operations become explicit lifecycles with visible state, retries, cancellation, and failure boundaries.

Keep authority on the backend.

The interface can guide a user, but security, state transitions, and system rules belong where they can be enforced.

Validate before you spend.

Reject malformed or unsafe input at the edge before it reaches an expensive or sensitive execution path.

Verify from the outside in.

Unit tests prove isolated behavior; integration and browser tests ask whether the whole thing behaves as a user sees it.

I like systems that leave fewer surprises behind them.

03 / STACKTOOLS WITH DIRECT EVIDENCE

Broad enough for the product. Centered enough to be believable.

BACKEND

Python · FastAPI · Flask · Pydantic · SQLAlchemy · Alembic

APIs / validation / persistence

SYSTEMS

PostgreSQL · Redis · ARQ · WebSockets · FFmpeg · Uvicorn · Docker

state / jobs / realtime / processing

FRONTEND

React · Next.js · TypeScript · Vite · Tailwind CSS

product interfaces / dashboards

VERIFICATION

Pytest · integration tests · Playwright · GitHub Actions

behavior / browser / automation

FILTER / SOURCE

Not every tool on a resume deserves equal weight. These are the ones with direct evidence in the selected systems.

04 / CONTACTOPEN CHANNEL

Bring a real problem.

I’m looking for backend / full-stack engineering opportunities and technical project conversations where the work has substance.

DIRECTsaiyedfaizan842@gmail.com ↗

GitHub ↗LinkedIn ↗

Email opens your mail client. No third-party form service.