Full-stack product engineering
Responsive, accessible front ends (React, TypeScript) wired to well-designed APIs — shipped end to end.
Currently at Capital One · Plano, TX
Senior Software Engineer.
Full-stack, cloud, data & AI.
I'm Brian Ta. A senior engineer who ships across the whole stack — from polished front ends and Java / Spring Boot services to databases, cloud infrastructure, and data & AI pipelines — for payments and banking systems at Capital One, Square and JPMorgan Chase.
- 📍 Dallas–Fort Worth, TX
- 💼 8+ years in fintech
- 🧱 Full-stack · Cloud · Data & AI
- 🎓 UT Dallas
8+years building production fintech systems
Full-stackfront end, back end, and everything between
Cloud + DataSQL & NoSQL, AWS / Azure / GCP, pipelines
AI-readydata engineering and applied ML / LLM features
In a hurry? Pick who you are.
- Role: Senior Software Engineer — full-stack, cloud and data.
- Strongest at: Java / Spring Boot, plus modern front end, databases, cloud and data/AI engineering.
- Domain: Fintech — banking at Capital One and JPMorgan Chase, payments at Square.
- Location: Plano / Dallas, TX. Open to hybrid or remote conversations.
- Education: The University of Texas at Dallas.
- Java / Spring Boot services and React / TypeScript front ends that share clean, typed contracts.
- Right database for the job: relational, document, key-value, search and analytics stores.
- Cloud-native delivery on AWS, Azure and GCP with containers, IaC and CI/CD.
- Data pipelines, warehouses and ML / LLM features built on solid engineering foundations.
- Small PRs, honest code reviews, and docs that new teammates can follow.
- One engineer who can take an idea from UI to database to cloud — fewer handoffs, faster delivery.
- Experienced with PCI, security reviews, audit trails and regulated environments.
- I communicate trade-offs plainly — cost, speed, risk — so decisions are easy.
- Comfortable owning a problem end to end, from idea to on-call.
What I do
Backend & microservices
Java / Spring Boot is my sharpest tool, alongside Node.js and Python, for resilient REST, gRPC and event-driven services.
Databases & data modeling
PostgreSQL, MySQL, Oracle, MongoDB, DynamoDB, Redis and Elasticsearch — schema design, tuning and migrations.
Cloud & DevOps
AWS, Azure and GCP; containers, Kubernetes, Terraform, CI/CD and cost-aware architecture.
Data engineering
Batch and streaming pipelines with Kafka, Spark and Airflow feeding warehouses and analytics.
AI & ML integration
LLM-powered features, retrieval (RAG), embeddings and model-serving integrated safely into production systems.
Selected work
Representative problem areas from my career. Details are kept general to respect employer confidentiality.
Payment authorization & processing service
Problem: Payment volume had to grow without sacrificing correctness or latency.
Built: Idempotent Java / Spring Boot services with strict transactional consistency and safe retries.
RequestValidateAuthorizeLedger
Merchant & customer dashboards
Problem: Users needed fast, clear visibility into money movement.
Built: React / TypeScript front ends over typed APIs with real-time updates and accessible design.
QueryAPIRenderUpdate
Fraud & risk scoring pipeline
Problem: Risk teams needed early, explainable signals.
Built: Streaming features and ML-scored decisions feeding rules with full audit trails.
IngestFeaturesScoreAlert
Cloud migration & infrastructure as code
Problem: Legacy infrastructure was slow to change and costly to run.
Built: Containerized services on AWS with Terraform, autoscaling and automated deployments.
PlanProvisionDeployScale
Ledger & settlement engine
Problem: Money movement needed a trustworthy, auditable source of truth.
Built: Double-entry ledger with immutable entries, balance checks and settlement jobs.
PostBalanceSettleAudit
LLM assistant with retrieval (RAG)
Problem: Internal teams lost time searching docs and runbooks.
Built: Embedding-based search plus an LLM assistant with guardrails, citations and evaluation.
EmbedRetrieveGenerateEvaluate
Data platform & analytics warehouse
Problem: Teams couldn't trust or easily query their data.
Built: Airflow-orchestrated ELT into a warehouse with quality checks and lineage.
ExtractLoadTransformServe
Internal tools & admin portals
Problem: Ops teams relied on manual scripts and spreadsheets.
Built: Self-service web apps with role-based access, workflows and audit logging.
AuthWorkflowApproveAudit
Event streaming backbone
Problem: Tightly coupled services made changes risky.
Built: Kafka-based events with the outbox pattern, schemas, replay and dead-letter queues.
ProduceStreamConsumeReplay
CI/CD & developer experience
Problem: Slow, manual releases made delivery risky.
Built: Automated build, test and deploy pipelines with quality gates and safe rollouts.
BuildTestScanDeploy
Observability & SLOs
Problem: Incidents were noisy and slow to diagnose.
Built: Distributed tracing, dashboards and SLO-based alerting tied to customer impact.
DetectTriageFixLearn
Database & performance tuning
Problem: Peak traffic stressed critical queries and services.
Built: Query and index tuning, caching, load testing and graceful degradation.
MeasureTuneVerifyShip
How I work
Understand the problem
I start with the real goal and constraints before choosing any tool or pattern.
Ship in small steps
Incremental delivery, feature flags and fast feedback keep risk low and momentum high.
Own it in production
I measure, monitor and improve — the work isn't done until it runs well.
Experience
-
Capital One
Jun 2026 — Present- Build full-stack banking features: React front ends, Java / Spring Boot microservices and the data layer behind them.
- Design cloud-native architectures on AWS with containers, IaC and automated delivery.
- Apply data and AI engineering — pipelines, analytics and ML-assisted features — to real product problems.
- Lead design reviews, define API contracts and mentor engineers on quality and resilience.
-
Square
Jun 2021 — Feb 2026- Built and operated payments services handling high volumes of merchant transactions, with Java / Spring Boot at the core.
- Delivered web interfaces and internal tools end to end, from UI through API to database.
- Designed idempotent, fault-tolerant, event-driven flows using Kafka and transactional outbox patterns.
- Worked across SQL and NoSQL stores and data pipelines powering reporting, risk and analytics.
- Improved reliability with observability, alerting and automation across the cloud platform.
-
JPMorgan Chase & Co
Feb 2019 — Jun 2021- Developed secure, compliant full-stack banking applications with Java / Spring backends and modern web front ends.
- Built REST APIs and integrations with internal and third-party systems.
- Worked extensively with relational databases — schema design, query tuning and migrations.
- Automated manual processes and strengthened unit, integration and release testing.
Education
-
The University of Texas at Dallas
2015 — 2018
Toolbox
Back end
JavaSpring BootSpring CloudSpring SecurityNode.jsPythonRESTgRPCGraphQL
Front end
ReactTypeScriptJavaScriptAngularNext.jsHTML / CSSTailwindAccessibility
Databases
PostgreSQLMySQLOracleSQL ServerMongoDBDynamoDBRedisElasticsearch
Cloud & DevOps
AWSAzureGCPDockerKubernetesTerraformJenkinsGitHub Actions
Data engineering
KafkaSparkAirflowdbtSnowflakeBigQueryETL / ELT
AI & ML
LLMsRAGEmbeddingsVector DBsscikit-learnPyTorchModel serving
Quality & observability
JUnit 5MockitoTestcontainersCypress / PlaywrightMicrometerGrafanaTracing
Practices
System designMicroservicesEvent-drivenSecurityAgileCode reviewMentoring
Let's talk
Open to conversations about engineering, systems, and new opportunities.