Role Backend Systems Developer
Location Lagos, Nigeria
Focus Concurrency · Job Queues · Geo-spatial · Marketplace APIs
GitHub @Falasefemi2 →
Status Open to work

01 — Systems Engineer / Profile

Femi.

Backend Systems & Infrastructure Engineer

Working stack TypeScript / Node · Java / Spring Boot · Bun · PostgreSQL · Redis · PostGIS · Docker · Prisma · Drizzle

02 — Signature Moment

Live Queue & Pipeline Simulator

Interactive execution runner comparing atomic database polling against concurrent fiber processing streams.

Engine Paladin
Workers 0
Queued 0
Execution mode

Jobs Table · Ingestion Pool

No pending jobs in DB. Click Enqueue Test Job to populate.

Concurrent Workers · SELECT FOR UPDATE

WORKER_01
State: IDLE
WORKER_02
State: IDLE
WORKER_03
State: IDLE

Lock Retries / Schedules

No active locks or retry schedules.

03 — Selected Work

Open-source backend systems, libraries, and marketplace components.

Code design focused on concurrency control and type safety.

  1. 01 Library

    Paladin

    import { PaladinQueue } from "@falasefemi2/paladin"

    Problem — Prevent race conditions and redundant execution when multiple distributed workers claim jobs simultaneously without memory-locking bottlenecks or external coordination layers like Redis locks.

    Approach — A lightweight Postgres-backed job queue on a single jobs table. Workers claim entries atomically via SELECT FOR UPDATE SKIP LOCKED, securing targets without blocking other threads from subsequent work.

    →src/ →live_simulator/ 
    interface PaladinConfig {
  runtime: NodeJS <TypeScript>;
  database: PostgreSQL;
  concurrencyModel: "SKIP LOCKED";
  atomicClaimQuery: string; // SELECT FOR UPDATE SKIP LOCKED
  locksNeeded: false;
}
  2. 02 Platform

    EasyRent Platform

    import { EasyRentCore } from "@falasefemi2/easyrent"

    Problem — Coordinate complex spatial property querying (Lagos maps) while maintaining secure, non-blocking authentication session lifetimes for API integrations.

    Approach — Full-stack Nigerian rental platform. The backend integrates PostGIS for proximity coordinate matching, Redis cache boundaries, and Drizzle ORM. The frontend uses Next.js with a custom fetch engine built around Effect-TS ManagedRuntime that auto-intercept-refreshes expired access tokens.

    →backend_src/ →frontend_src/ →api_docs/ →live/ 
    type EasyRentStack = {
  engine: Effect <v4, Bun>;
  database: PostgreSQL <PostGIS, Drizzle>;
  caching: Redis;
  authLayer: ManagedRuntime <TokenRefresh>;
  client: NextJS_15 <Tailwind>;
}
  3. 03 System

    VendorHub v2

    import { VendorHubEngine } from "@falasefemi2/vendorhub-v2"

    Problem — Small-scale WhatsApp micro-merchants experience extreme sales attrition because status catalog listings expire and disappear every 24 hours.

    Approach — A WhatsApp catalog-preservation backend. Uses Node/TypeScript, Express, and Prisma ORM to index status-media assets via Baileys API handlers, committing uploads to Postgres. Serves persistent static Next.js storefront pages so buyers can browse inventories indefinitely.

    →backend_src/ →frontend_src/ →live/ 
    interface VendorHubSystem {
  apiGateway: Express <TypeScript>;
  dataMapping: PrismaORM;
  importer: BaileysClient; // whatsapp crawler
  persistence: PostgreSQL;
  cacheState: Redis;
}
  4. 04 Service

    Cronforge

    import { CronforgeScheduler } from "@falasefemi2/cronforge"

    Problem — Scattered engineering articles on technical hubs require manual daily discovery, reading, and bookmarking.

    Approach — A daily scheduled parser in Node/TypeScript on Bun. Executes background cron schedules that fetch relevant article payloads from the Dev.to endpoint and delivers a formatted newsletter to user email via Resend integrations.

    →src/ 
    type CronforgeStack = {
  runtime: Bun;
  scheduler: NodeCron;
  orm: Drizzle;
  database: PostgreSQL;
  dispatch: ResendAPI;
}
  5. 05 Reference

    security-docker

    import { SecurityDocker } from "falasefemi2.security.docker"

    Problem — Configure robust stateless user access controls on Spring Boot enterprise applications while ensuring isolation across target production micro-deployments.

    Approach — A secure Java REST API demonstrating JWT-based authentication and authorization architectures. Custom Spring Security filter chains for token verification, isolated and compiled into a secure container deployment using Docker.

    →src/ 
    class SecurityDockerSetup {
  framework: SpringBoot <Java>;
  securityScope: SpringSecurity;
  authScheme: JWT;
  containerization: Docker;
  db: PostgreSQL;
}
  6. 06 Library

    paystack-effect

    import { PaystackClient } from "@falasefemi2/paystack-effect"

    Problem — Integrating Paystack's payment API inside functional-effect architectures traditionally relies on untyped fetch calls, manual parsing, and no compiler-enforced request/response payloads.

    Approach — A type-safe Effect-TS v4 wrapper for the Paystack API. Exposes all resource endpoints (Transaction, Customer, Subscription, Transfer, etc.) as clean Context.Service definitions behind reusable dependency Layers, verifying runtime schemas with fully-typed request/response validation.

    →src/ 
    interface PaystackEffectSdk {
  runtime: Effect <v4>;
  typeValidation: Schema;
  services: Context_Service <
    | "Transaction"
    | "Customer"
    | "Subscription"
    | "Transfer"
  >;
  architecture: Layer;
  clientType: "Library";
}
  7. 07 Example App

    paystack-effect-example

    import { PaystackExample } from "@falasefemi2/paystack-effect-example"

    Problem — Demonstrating full checkout payment flows securely requires coordinating price computation, transaction validation, and webhook signature verification under a production-grade backend and frontend.

    Approach — A runnable checkout reference application. The backend uses Bun and Effect v4 to handle server-side pricing computations and webhook signature verifications; the frontend uses Next.js 15 and shadcn/ui for a responsive product catalog and payment checkout interface.

    →src/ 
    type PaystackExampleStack = {
  frontend: NextJS_15 <shadcn_ui>;
  backend: Bun <Effect_v4>;
  integration: PaystackEffect;
  security: WebhookVerification;
  clientType: "Example App";
}
  8. 08 Gallery App

    Image Gallery

    import { ImageGallery } from "@falasefemi2/image-gallery"

    Problem — Organizing and searching high-resolution visual assets requires dynamic grid layouts, rapid image filtering, and non-blocking preview modals without sacrificing client-side render speed.

    Approach — A responsive web image gallery featuring multi-layout view modes, client-side asset filtering, and interactive full-screen modal previews built with performance-optimized image rendering.

    →src/ →live/ 
    interface ImageGallerySystem {
  layoutEngine: MasonryGrid <Responsive>;
  filtering: ClientSideSearch;
  modalViewer: LightboxPreview;
  optimization: LazyLoading <SrcSet>;
  clientType: "Gallery App";
}

· Non-Public Enterprise / Core Implementations

  • effectq →src/

    Personal PostgreSQL job queue, pg-boss model, built on Effect-TS v4 + Bun. Uses Context.Service, Layer.scoped + acquireRelease, Stream-based workers with SKIP LOCKED polling, Schedule for retries/cron, and typed error unions.

  • Chowdeck Engine (Clone) →src/

    Backend core managing order lifecycles via deterministic state machines. Stack uses Drizzle, @effect/sql-pg, background dispatch with forkDetach, and spatial proximity calculations.

  • EasyWorka →src/

    Enterprise HR application built solo at Vatebra. Centerpiece is a generic approval-workflow state machine covering Leave, Exit, Memo, Voucher, and Trip approvals.

04 — Tooling

Core languages, runtimes, and infrastructure tooling used in production.

TypeScript
Node.js
Bun
Java
Spring Boot
PostgreSQL
Redis
Docker
Nginx
Neovim

05 — Experience

Execution logs and software contributions across enterprise and agency environments.

  1. June 2025 – Present

    Vatebra Limited

    Frontend Developer
    • Delivered multiple frontend projects, including the Nigeria Computer Society website and an internal HR platform used company-wide.
    • Built and shipped a range of client and internal-facing web applications as part of a fast-paced, multi-project frontend workload.
    • Worked independently across projects, owning frontend implementation from requirements through deployment.
  2. Sept 2024 – Jan 2025

    Noemdek Limited

    Frontend Developer
    • Built and maintained frontend for an advertising web application, handling UI implementation and feature development.
    • Developed frontend for a transportation web application, translating product requirements into functional interfaces.
    • Worked across two concurrent products, adapting to different codebases, design patterns, and business domains.
  3. Jan 2021 – Mar 2021

    Sterling Bank

    Frontend Developer Intern
    • Internship focused on foundational frontend development, building hands-on experience with core web technologies under senior engineer guidance.
    • Worked on internal frontend tasks and exercises to strengthen practical skills in HTML, CSS, and JavaScript fundamentals.
    • Gained exposure to enterprise development workflows and coding standards within a banking technology environment.

06 — Curriculum

Systems CV / Resume

Femi's engineering curriculum — professional experience at Vatebra and backend infrastructure projects — documented in detail.

↗ Open Resume Doc

07 — Diagnostics & Contact

Available for backend systems design, database optimizations, REST API development, and distributed worker integrations.

GitHub
github.com/Falasefemi2
Email
femifalase228@gmail.com
Status
Available for hire