Yoandry Collazo Barroso

Senior Software Engineer · LLM Platform Engineering

Profile

I build the engineering around large language models, the part that decides whether an LLM feature is safe to put in front of customers. At NDAX, a Canadian crypto exchange, that has meant building an AI platform end to end: agent orchestration, a layered guardrail stack, a replayable eval harness, an authenticated gateway between models and customer accounts, and an MCP tool layer over internal services. I also lead the team that builds on it. Behind the AI work sit more than ten years of production systems engineering in data centres, fintech, aviation and e-commerce, which is why my AI work looks like systems engineering: contracts, tests, observability and an honest account of the failure modes. In a regulated environment, “the model said so” is not an acceptable answer to a compliance reviewer, and designing for that constraint from the start is most of the work.

LLM platform engineering Agent orchestration RAG & grounding LLM evaluation Guardrails & AI safety Model Context Protocol Regulated finance

Experience

Jun 2025 – Present Calgary, AB

NDAX·Senior Software Engineer

Contract from Jun 2025 · full-time since Jan 2026

Backend engineering across the exchange’s services from June 2025, and its AI platform since January 2026. I own the architecture across the platform’s three services, review the changes that go into them, and lead the AI integrations team. Over 1,100 pull requests across 22 repositories since September 2025.

  • Agent orchestrator. Led development of the service that turns a customer message into a grounded answer: intent routing, prompt assembly with few-shot selection, tool calling against internal APIs, streamed responses over WebSockets, session continuity and erasable user memory. TypeScript, Azure AI Foundry, Redis, Socket.IO; over 250 test files.
  • Guardrails. Designed the response-safety layer: 16 independent checks covering groundedness, structured-output validation with fallback, formatting integrity, provenance and compliance-sensitive content. Replaced embedding-similarity grounding with an entailment-based LLM judge after showing that similarity both false-blocked safe replies and passed answers the corpus did not support.
  • Eval harness. Built the in-repo evaluation framework (runner, assertion library, LLM judge, versioned baselines), so a prompt or model change is measured against a replayable suite.
  • AI gateway. A NestJS service that is the single authenticated ingress for AI traffic and the only path to exchange internals, with over 140 endpoints. Internal APIs reach agents as typed tools with generated parameter schemas, per-tool access control and full audit logging.
  • MCP tooling. Model Context Protocol servers that expose OpenAPI specs and internal service operations as tools over STDIO and HTTP, under the APIs’ own access controls. In production for internal testers.
  • Internal developer tooling. Wrote ndax-pilot, a Go CLI coding agent running on the company’s own Azure AI Foundry deployments, with a context engine, persistent repository memory and a skill system, so engineers get agentic assistance without code leaving our tenancy.
  • Team lead, AI integrations. Lead two junior engineers. I write every ticket they pick up to a ready-for-work standard (one repository, one pull request, one working day, no open design decision) and run weekly sprints with estimates calibrated against measured cycle time. Security review is mandatory on authentication, money-movement and personal-data changes.
  • AI coding agent tooling. Built the team’s Claude Code plugin marketplace, which carries our ticket, estimation and review process into our coding agents. Releases ship only after a behavioural eval suite passes, and agents ask a person before writing to the issue tracker or code host.
  • Technical direction. Own the cross-service contracts (WebSocket events, HTTP headers, payload shapes, schema) and write the incident post-mortems, runbooks and development workflow the engineering team works to.
  • Exchange platform. Built a multi-tenant admin portal for business clients from scratch (Next.js in English and French, NestJS backend with four-eyes approvals). Reworked customer statement generation, and brought a payments service under strict TypeScript and CI.
Aug 2022 – Jun 2025

Quantifi / Credit App·Senior Software Developer

Joined Quantifi as Full-stack Software Developer, Aug 2022 · promoted to Senior in 2023 at the Credit App spin-out

Auto-lending finance. Event-driven platform engineering: event-sourced microservices on Kurrent (formerly EventStore), treating the event ledger as the system of record, built as Node.js services in an Nx monorepo.

Jan 2018 – Aug 2022 Miami, FL · remote

Front10·Lead Software Engineer

Full-stack Developer, Jan 2018 – 2019 · promoted to Lead 2019

Led the software development team building embeddable products for the airline industry.

  • Airline widgets. React components embedded on carrier websites (booking forms, fare maps and related flows), shipped for Spirit, Air Canada and LATAM.
  • Test engineering. Worked with the QA team to build the widgets’ coverage across unit, integration and end-to-end tests (Cypress).
  • Accessibility. Completed a full accessibility certification and led accessibility QA for the widget suite.
Jan 2017 – Jan 2018 Barcelona, Spain

LUPO Barcelona·Full-stack Developer

E-commerce platform engineering: led the store’s migration from WordPress to PrestaShop, and ran the DigitalOcean infrastructure alongside it. PHP, MySQL, Python.

Sep 2012 – Dec 2016 Havana, Cuba

Dirección de Tecnologías y Sistemas (DTS)·Systems Engineer, Data Centres

Data-centre systems engineering: Oracle database administration on RAC clusters, with GoldenGate replication and standby (Data Guard) failover, alongside Linux server administration across the estate.

Skills

Programming
TypeScript · JavaScript · Python · Go · SQL
AI & LLM
Azure AI Foundry · Model Context Protocol · agent orchestration · RAG and grounding · LLM-as-judge evaluation · guardrail design · prompt engineering
Backend
NestJS · Node.js · GraphQL · WebSockets / Socket.IO · REST and OpenAPI · Redis
Event-driven
Event sourcing · Kafka · Kurrent (formerly EventStore) · microservices · Nx monorepos
Frontend
React · Preact · Vite
Platform
Docker · Kubernetes · Terraform · Azure · AWS
Spoken
Spanish (native) · English (professional)

Education & Certification

2012 Havana, Cuba

Ingeniero en Ciencias Informáticas

Engineering in Informatics Sciences, Universidad de las Ciencias Informáticas (UCI).

Assessed by WES (2023) as equivalent to a Canadian four-year Bachelor’s degree.

2015 Shenzhen, China

HCNA · HCNP·Huawei Certified Network Associate & Professional

Data-centre technologies track. Huawei University, Shenzhen. Since renamed HCIA · HCIP.