Open to new opportunities

Kevin Ng

Senior Backend-Focused Full Stack Engineer

Austin, TX 78704

10+ years delivering scalable APIs, distributed systems, and cloud-native backends. Deep hands-on experience with AI/ML, LLMs, RAG pipelines, and blockchain infrastructure in production.

GoPythonNode.jsAWSKubernetesBlockchainRAG / LLMsKafka
10+Years Experience
3Companies
4Domains
Problems Solved

About Me

Building systems that scale & last

I'm a senior backend-focused full stack engineer with 10+ years of experience delivering production systems across healthcare, fintech, real estate, and e-commerce at companies including Thoughtworks, Kraken, and BigCommerce.

I own end-to-end delivery—from architecture and data modeling to shipping and postmortems—while staying comfortable partnering with front-end, product, and security stakeholders. My recent work spans integrating LLMs and RAG pipelines into production, owning high-throughput blockchain deposit infrastructure, and tuning database performance under extreme load.

I write clean, testable code, lead architecture reviews, and mentor engineers on scalable design patterns. I thrive in remote-first, async-heavy environments and care deeply about reliability, observability, and thoughtful system boundaries.

75%Doc processing time cut
80%RAG accuracy gain
60%Latency reduction
45%Tx bottleneck cut
Traffic spikes handled
30K+Enterprise users served

Backend Architecture

Microservices, REST & gRPC APIs, event-driven design, idempotency, and distributed systems at scale.

AI & Machine Learning

RAG pipelines, LLM APIs, vector databases (Pinecone, Weaviate), and reliable ML data pipelines in production.

Cloud & DevOps

AWS (S3, Lambda, EC2), Kubernetes, Docker, and CI/CD pipelines for multi-environment deployments.

Performance & Reliability

Observability with Prometheus, Grafana, Datadog, ELK; distributed tracing; proven track record of reducing latency.

Core Skills

The toolkit

⌨️

Programming Languages

GoPythonNode.jsJavaPHP
🏗️

Backend & Architecture

MicroservicesREST APIsgRPCGraphQLEvent-Driven DesignFastAPIFlaskSpring Boot
☁️

Cloud & DevOps

AWS S3AWS LambdaAWS EC2KubernetesDockerCI/CD Pipelines
🗄️

Data Stores

PostgreSQLMySQLRedisMemcached
📨

Messaging & Streaming

Apache KafkaRabbitMQGoogle Pub/Sub
🤖

AI & Machine Learning

RAGLLM APIsPineconeWeaviateVector DatabasesAI Data Pipelines
📊

Observability

PrometheusGrafanaDatadogELK StackDistributed Tracing
🔐

Security & APIs

OAuth 2JWTRBACOWASPIdempotencyTDDDDD

Career

Professional Experience

  • Led architecture and design reviews for client programs; aligned engineering plans with business priorities, delivery constraints, and budget.
  • Designed and delivered an AI-powered healthcare data pipeline that cut manual document processing time by 75% for medical provider workflows.
  • Shipped a RAG solution that raised accuracy of extracted medical insights by 80%.
  • Ran asynchronous ingestion on Python and AWS Lambda, supporting thousands of concurrent document uploads per day.
  • Improved a real estate recommendation service by 60% end-to-end latency through Redis caching, query design, and layered cache patterns.
  • Tuned PostgreSQL for high-volume property listings and cut database lock contention by 40% through indexing and query changes.
  • Designed chunk-level idempotency for processing pipelines so partial failures did not create duplicate records.
  • Co-led retrieval and data-model design with ML engineers; raised vector search relevance by ~30% through indexing improvements.
  • Streamlined CI/CD workflows for AI services, cutting deployment frequency from weekly to daily.
GoPythonFastAPINode.jsAWS LambdaPostgreSQLRedisRAGPinecone

Portfolio

Projects & Impact

Production systems built across healthcare, fintech, real estate, and enterprise e-commerce — each with measurable outcomes.

Healthcare AI Pipeline — design preview
Featured · Healthcare

Thoughtworks × Medical Provider Client

AI-Powered Healthcare Data Pipeline

Built a Python and FastAPI pipeline leveraging LLMs to structure unstructured medical notes with ~80% higher accuracy. Added retrieval-augmented generation (RAG) with Pinecone for accurate extraction across thousands of documents per day.

75%

Less manual processing time

80%

Higher extraction accuracy

85%

Fewer processing failures

PythonFastAPILLMsRAGPineconeAWS LambdaPostgreSQL
Real Estate Predictive Engine — design preview
Featured · Real Estate

Thoughtworks × Real Estate Platform

Real Estate Predictive Engine

Built a Node.js and Python backend serving personalized property recommendations from real-time user behavior signals. Lowered recommendation latency by 60% with asynchronous, event-driven processing and a layered Redis cache.

60%

Latency reduction

40%

Less DB lock contention

20%

User engagement lift

Node.jsPythonRedisPostgreSQLREST APIEvent-Driven
Fintech

Blockchain Deposit Infrastructure

Kraken — Cryptocurrency Exchange

High-throughput Go and Kafka pipeline for real-time cryptocurrency deposit detection and ledger attribution. Idempotent, versioned ledger updates tolerated out-of-order events with 100% data integrity.

Traffic spikes absorbed

45%

Tx bottleneck cut

100%

Data integrity maintained

GoApache KafkaKubernetesPostgreSQLRedisDistributed Systems
Enterprise

Enterprise Session Management System

BigCommerce — E-Commerce Platform

High-availability authentication stack in Java and Spring Security for 800+ concurrent logins/minute. Hierarchical RBAC and audit logging for 30,000+ enterprise users; eliminated 95% of session cache race conditions.

800/min

Concurrent logins

30K+

Enterprise users

95%

Fewer surprise logouts

JavaSpring BootMySQLMemcachedOAuth 2JWTRBAC

Education

Academic Background

Bachelor of Science in Computer Science

The University of Texas at Austin

Austin, TXMay 2011 – May 2015
Algorithms & Data StructuresDistributed SystemsDatabase SystemsSoftware Engineering

Get In Touch

Let's work together

Open to senior backend engineering, staff engineer, and technical lead roles. Remote-first. Available for a conversation about your next big challenge.

Ready to connect?

Whether you have a backend challenge, a technical leadership gap, or just want to talk architecture — I'm happy to chat.

Austin, TX 78704 · Available remotely worldwide