PTEN

Will Peixoto with arms crossed, black t-shirt with the orange lambda

Will Peixoto

São Paulo, Brazil

Serverless, AWS architecture and AI agents: what I'd pick, why, and what it costs.

Staff Engineer at Serverless Guru. My topics usually come from a technical problem I had to solve or something I had to explain.

BUILD. SCALE. REPEAT.

focus
Serverless, architecture and AI agents on AWS
languages
Portuguese and English, on the blog and on stage
meetups
One of the leaders of AWS User Group São Paulo, 120+ meetups
events
Co-organizes ServerlessDays São Paulo
stage
Speaker and mentor at the Serverless + Agentic AI na AWS workshop (Sem Servidor, 2026)
Read the blog Invite me to speak

resiliencePT · EN

Featured, series Architecture that Survives

Is multi-cloud more resilient? The math says no

  1. Context

    After every big cloud outage, the pitch comes back: “Let's go multi-cloud, so if AWS goes down we stay up on Google.”

  2. Rule of thumb

    Multi-AZ and multi-region on the same cloud first. Multi-cloud only for a business reason, and then active-passive, documented in an ADR.

  3. Trade-off
    What you gain

    Native managed services, the same IAM and the same tooling, for a fraction of the complexity.

    What you pay

    You stay exposed to the very rare event of the whole provider going dark.

  4. Would change if

    Regulation, a contract or data sovereignty requires a second cloud.

Read the full post

Decision log

Every post ends in an architecture decision: the context, what I'd pick, what you gain and what you pay.

  1. streamingPT · EN

    Data Streaming on AWS: Kinesis, Firehose, Flink, or MSK?

    Rule of thumb: Kinesis when there's no commitment to Kafka; MSK when the team already lives on Kafka.

    Context and trade-off

    Four streaming services on AWS, and “which one is best” is the wrong question: the choice comes down to context and ecosystem.

    What you gain

    Kinesis is simpler and native, and fits better with Lambda and the rest of serverless.

    What you pay

    Ordering is per shard, not global, and at-least-once delivery needs an idempotent consumer.

  2. lambdaPT · EN

    AWS Lambda MicroVMs: run untrusted code with VM-level isolation (no infra to manage)

    Rule of thumb: Lambda Functions for the event-driven backbone; MicroVMs only for the steps that run untrusted code.

    Context and trade-off

    Container, VM or Lambda Function: none of them gave VM isolation, fast launch and per-session state at the same time.

    What you gain

    VM isolation with near-instant launch and state kept for up to 8 hours per session.

    What you pay

    No sa-east-1 yet, ARM64 only, capped at 16 vCPU and 32 GB.

  3. resiliencePT · EN

    High Availability Has a Price: Resilience Is a Decision, Not a Stack

    Rule of thumb: Choose between HA, multi-region and on-premise by the cost of each nine and by RTO/RPO, and record the accepted risk in an ADR.

    Context and trade-off

    Resilience is a business decision before it's a technical one, and the context behind each choice gets lost if nobody writes it down.

    What you gain

    The risk is mapped, owned and justified. Failure doesn't catch the team by surprise.

    What you pay

    Not everything gets HA: batch jobs, internal systems and dashboards can tolerate minutes or hours down.

  4. changed Aug 2026AI agentsPT · EN

    AWS MCP Server: which one to use, when, and how to set it up (now that AWS recommends just one)

    Rule of thumb: One server only, the managed AWS MCP Server. Knowledge is for those who just want to query docs, without an AWS account.

    Context and trade-off

    When the post came out, the answer was “it depends on the pain”. Three weeks later, the official setup guide started recommending a single server: duplicated tools confuse the agent.

    What you gain

    Less config, fewer conflicts, the same IAM leash.

    What you pay

    An agent that operates the account: start in a sandbox, with scoped IAM and a human in the loop for irreversible actions.

See all posts

Community

Community is half the fun of this career. If you're starting out or want a stage for your first talk, come to a meetup and say hi.

AWS User Group São Paulo

one of the leaders

AWS meetups in São Paulo, past 120 of them. If you want a stage for your first talk, this is the place.

See upcoming meetups

ServerlessDays São Paulo

co-organizer

The serverless community event in São Paulo.

About the event

Talks

Talk, workshop, mentoring or panel, in Portuguese or English. Invitations reach me fastest by LinkedIn DM.

Sem Servidor, Florianópolis, Brazil, 17 Oct 2026

Serverless + Agentic AI na AWS

in-person workshop, talk and hands-on mentoring

Topics I speak on

  • Lambda durable functionslong-running workflows as plain sequential code, and when not to use them
  • Event-driven in the age of AI agentslevel 400what breaks in retries, schemas and loops when the consumer reasons
  • Serverless and event-driven on AWSpatterns, resilience and cost trade-offs in production
  • AI agents with serverless on AWSlevel 100from an LLM on Amazon Bedrock to an agent with Strands running on Lambda
  • AI agents and MCP on AWSidentity, governance and least privilege for agents
  • Lambda MicroVMsisolation for untrusted code
Invite me on LinkedIn See talks

Where to find me

Everything I publish starts here on the blog. Elsewhere you'll find the cross-posts, the conversations and new post alerts.

New posts, straight to your inbox

One serverless, architecture or AI agents topic on AWS per issue: what I'd pick, why, and what it costs. No spam.

Opens the subscription page in a new tab.