Ervin Halmi

Available to hire

Skills

Experience Level

Python
Expert
Django
Expert
Docker
Expert
Kubernetes
Expert
JavaScript
Expert
AWS
Expert
SQL
Expert
Spark
Expert
React JS
Expert
PyTorch
Expert
PostgreSQL
Expert
Node.js
Expert
HTML5
Expert
Google Cloud
Expert
API
Expert
MySQL
Expert
Redis
Expert
TensorFlow
Expert
TypeScript
Expert
Keras
Expert
Neo4j
Expert
GraphQL
Expert
GitHub Copilot
Expert
Elasticsearch
Expert
DynamoDB
Expert
GIS
Intermediate
See more

Language

English
Fluent

Work Experience

Senior Software Engineer at Google
January 1, 2023 - Present
Led a cross-functional backend team delivering production-grade microservices and real-time APIs, scaling to thousands of active users with 99.99% uptime. Architected AI integrations using Vertex AI and OpenAI APIs with secure API key management, message routing, and feedback-loop accuracy monitoring, improving automated support resolution by 40%. Built high-availability data ingestion pipelines (Airflow, Spark, Pub/Sub, BigQuery), reducing data freshness lag by 65% and cutting data loss with DLQs and automatic replay. Reduced backend latency by 45% through asynchronous FastAPI patterns, Redis caching, and query tuning. Productionized RAG chat systems with Pinecone embeddings and vector search, boosting assistant adoption. Implemented DevOps automation (Docker, Kubernetes, GitHub Actions) enabling 3x faster deployments with zero-downtime rollouts. Mentored engineers and ensured high-quality releases with testing and observability.
Full Stack Engineer at Cognizant
January 1, 2020 - January 1, 2023
Developed a national UAE rental exchange platform using Python + React + GIS, delivering real-time property availability, rental history, and pricing across major cities. Engineered geospatial features (OpenLayers, Mapbox, PostGIS) with spatial indexing for near-instant interactive maps. Built geospatial ETL pipelines; integrated AI/LLM services (OpenAI) for market analytics, automated pricing estimation, and property classification. Launched intelligent rental chatbot increasing self-service adoption; utilized Pinecone for vector search to match neighborhoods and predict rental trends. Implemented secure APIs with OAuth2; optimized backend with Redis caching, async I/O, PostGIS indexing; Built CI/CD pipelines; Migrated spatial workloads to managed cloud.
Software Engineer at Capgemini
January 1, 2017 - January 1, 2020
Implemented OCPP 1.6J communication services in Node.js for charging stations; developed EMS/CMS microservices; implemented secure payments and PCI tokenization; built real-time reporting and alerting; optimized Node.js performance with event-driven streams, caching, and async execution. Established microservices DevOps practices with Docker, container-based deployments, and environment automation. Collaborated with hardware teams and external CPO/OEM partners to validate OCPP compliance and charger compatibility. Created fault-tolerant reporting & alerting with anomaly detection and escalation workflows; contributed to frontend development using React with real-time dashboards.
Backend Developer at NTT DATA
January 1, 2014 - January 1, 2017
Developed RESTful APIs using Node.js + Express.js and Python (Django/FastAPI) to support authentication, billing, and reporting across multiple web products. Built and optimized PostgreSQL/MySQL schemas, indexes, and stored routines to improve query performance and data integrity. Implemented JWT-based auth, role permissions, and secure API request flows. Integrated third-party APIs and created reusable service modules. Developed asynchronous background jobs (Celery/Redis and BullMQ/Kafka) to improve responsiveness. Wrote unit and integration tests and contributed to CI/CD pipelines.

Education

Bachelor’s Degree in Computer Science at The University of Tokyo
April 1, 2008 - April 1, 2012

Qualifications

Add your qualifications or awards here.

Industry Experience

Software & Internet, Professional Services, Media & Entertainment
    Scalable AI-Powered Enterprise Platform

    Designed and delivered a production-grade full-stack platform using React, TypeScript, Python, FastAPI, Node.js, PostgreSQL, and AWS.

    I owned features end-to-end, from architecture and API design through frontend implementation, testing, deployment, and production monitoring. The platform used microservices, REST/GraphQL APIs, asynchronous processing, Redis caching, and event-driven workflows to support high-volume workloads.

    I also integrated LLM and RAG capabilities, including semantic search, vector retrieval, structured AI workflows, access controls, and monitoring.

    Key focus areas included scalability, API performance, reliability, security, automated testing, CI/CD, and cloud deployment. I worked closely with product and engineering teams, reviewed code, mentored developers, and helped drive technical decisions from design through production.