Samuel Oshin — Python Backend Engineer

Python Backend Engineer

AI Backends • Scalability

Award-winning Python Backend Engineer (2× HNG Finalist, Top 1–5% of 10,000+). Specializing in RAG pipelines, vector search, and event-driven architectures. 16× performance gains. Zero-fund-loss transaction systems. Production-grade AI backends.

Experience

Where I've made an impact

2023

First Dev Role

2024

NYSC + Open Source

2025

2× HNG Finalist

Now

Emerj LLC (AI)

From first code to AI-powered backends — building with purpose.

Dec 2025 – Present

Backend Engineer

Emerj LLC (Remote)

Details

Architecting Legal Watch Dog — an AI-powered regulatory monitoring platform processing 1,000+ global regulatory updates daily with a high-performance distributed pipeline.

Defined complete system architecture by authoring the Technical Requirements Document (TRD) and designing the ERD for multi-tenant scalability.
Engineered a high-performance 4-stage distributed pipeline for real-time web scraping and semantic analysis using FastAPI, Celery, and Redis.
Implemented robust concurrency control through Redis distributed locks and SQLAlchemy transactional integrity (FOR UPDATE SKIP LOCKED).
Reduced manual document review time by ~40% with AI-driven semantic change detection leveraging LLMs (OpenRouter/OpenAI/Gemini).
Optimized storage and processing costs by 30% with custom SHA-256 content deduplication and MinIO high-availability storage.
Engineered secure authentication with JWT flows and multi-provider OAuth (Google, Microsoft, Apple) integrations.

PythonFastAPICeleryRedisPostgreSQLOpenAIMinIODocker

Oct 2025 – Dec 2025 (HNG13) & Jan – Apr 2025 (HNG12)

Backend Python Engineer Intern

HNG Internship (Remote) — 2× Finalist

Details

Awarded Mentor's Choice — Best Backend Intern (HNG13). Ranked Top 5% of 10,000+ developers (HNG12). Built production-grade payment, RAG, and notification systems.

Achieved zero fund loss for the Wallet Service with atomic transactions, idempotent webhooks, and background recovery jobs using Redis.
Improved retrieval precision by ~25% for internal RAG pipeline with advanced document chunking, OpenRouter embeddings, and hybrid vector search in Pinecone.
Scaled notification throughput to 1,000+/min with 99.5% delivery rate using RabbitMQ, circuit breakers, and dead-letter queues.
Reduced data endpoint latency by 16× (156s → 9.6s) by refactoring to asyncio parallelism and bulk SQL updates (ON DUPLICATE KEY UPDATE).
Built 'PRRover', an AI tool using A2A protocol and Google Gemini 2.0 for automated security reviews with sub-second code analysis.

PythonFastAPIRedisRabbitMQPineconeGoogle GeminiCeleryDocker

Aug 2024 – May 2025

IT Support Officer (NYSC)

Coleman Technical Industries

Details

Improved system uptime by 25% by automating Oracle/Odoo ERP backups using custom Python scripts with scheduled cron jobs and error-handling.

Reduced downtime incidents by 20% by resolving 100+ user support tickets through systematic troubleshooting and Group Policy implementations.
Automated ERP backup processes for Oracle and Odoo systems using Python scripts with cron job scheduling.
Documented Standard Operating Procedures (SOPs) for Odoo application role rights and access management.
Implemented automation for adding new staff to Active Directory using Python scripts.

PythonOracleOdooActive Directory

Nov 2023 – Apr 2024

Software Developer

WML-Integrated Solutions @ PZ Cussons

Details

Reduced IT issue resolution time by 40% by developing a custom Django-based ticketing system with automated workflows and real-time notifications.

Built a custom Django-based ticketing system with automated workflows, reducing IT issue resolution time by 40%.
Prevented 15+ potential network outages by building internal Python monitoring tools with proactive alerting and data logging.

PythonDjangoPostgreSQL

Core Technologies