Claude Certified Architect — Foundations

I architect production AI systems on Claude: agents, MCP servers and data platforms, from architecture to production.

01 — About

Software businesses actually run on.

experience

eCommerce platforms, multi-tenant B2B systems, analytics platforms, and now AI at scale.

20+ years
ai systems

Multi-agent orchestration, MCP servers and RAG pipelines on the Claude API and AWS Bedrock.

Claude · MCP
platforms

FastAPI services, Postgres, dbt semantic layers, SSO and API gateways.

FastAPI · dbt
credentials

Arts et Métiers engineer. Claude Certified Architect, by Anthropic.

Certified
02 — Principles

A product should be.

CLAUDE.md+6 −0
<CORE_PRINCIPLES>A product should be:- Beautiful and appealing- Focused on a few core features- Simple and solid by design</CORE_PRINCIPLES>
03 — Approach

From architecture to production.

i.

Architect

I own a project from requirements through production: design, build, code review and rollout.

ii.

Build

Agents, MCP servers and APIs, with auth, cost, latency and observability handled in production.

iii.

Lead

Technical direction, code review, and Claude Code adoption for existing engineering teams.

“He asks the right questions: he takes the time to understand the problem deeply before jumping to a solution.”
Suchisubhra Sinha — Enterprise Architect
04 — Selected work

Built for production, not for demos.

trossatanalytics
brief ›

Self-service BI for a US B2B fintech.

Scope

The full analytics stack of a B2B commerce platform: FastAPI services, Lightdash BI on dbt and semantic models, Databricks and Redshift.

Impact

Reporting moved out of legacy software. Business teams and clients reach their own data and dashboards directly.

Complexity

Entra ID SSO, row-level filtering for tenant isolation, analytics content managed as code.

An agent platform on Claude & AWS Bedrock.

Scope

Multi-agent orchestration behind a FastAPI gateway, exposing agents to every other service in the company.

Impact

Existing software gained new AI features and agents, with cost, latency and observability controls in production.

Complexity

Enterprise APIs exposed as agent tools through an MCP server, and as REST endpoints through API Gateway. SSO, rate limiting, request validation.

Revenue by region, last quarter?

# No raw SQL: the agent resolves the question
# into a governed dbt Semantic Layer query

query = {
  "metrics": ["revenue"],
  "group_by": ["customer__region"],
  "where": "metric_time >= '2026-04-01'"
}

# MetricFlow compiles the SQL. One definition
# drives the schema, the dashboards and the answer.

Assistants and extraction over company documents.

Scope

RAG assistants on Bedrock Knowledge Bases, and a multimodal agent extracting metadata from documents.

Impact

Answers grounded in company documents. Fields extracted on demand, with low-confidence results routed to a human.

Complexity

S3 vector storage, metadata-based tenant isolation, event-driven ingestion, per-request field specs and confidence scoring.

An autonomous agent running a BTC DCA strategy.

Scope

FastAPI + Claude agent optimizing a dollar-cost averaging strategy.

Impact

Fully automated, market-driven investment decisions, running 24/7.

Complexity

Three-layer strategy, sentiment-driven parameters, scheduling, two-way Telegram interface.

Let's build.

Short and long-term projects, remote from Andorra.
US LLC, invoiced in USD, on Eastern hours.

hello@trossat.com