Thiago Amaro
Solutions Architect · Mainframe Modernization & AI

Thiago Amaro

Seventeen years in mainframes: first running z/OS systems, then leading the technical side of presales for the enterprises that depend on them. These days I work on getting AI adopted in that world, and I build the tools in public.

About

I spent eleven years as a mainframe system programmer at IBM, Broadcom, and Software AG, the kind of work where a mistake shows up on the evening news. Then I moved to the other side of the table: technical presales at Kyndryl, helping large banks, insurers, and governments decide what to do with the systems everything else runs on.

That job changed over the last two years. Every modernization conversation turned into an AI conversation, and very few people in those meetings had spent real time on both sides.

So I built things. In the beginning it was all for learning, but then they started to get useful: a multi-agent AI system my presales colleagues used every day, and open-source tooling for Claude Code that genuinely helps me in my daily work.

It didn't stop at work. Claude helped coordinate my family's move from Prague to Madrid, and with a private server at home it now takes care of my home network, my home automation, and my two PCs.

Starting early meant there was little to copy, so I had to learn the primitives myself: context and memory management, skills and workflows, skeptic review rounds, loops that monitor things, loops that improve the loops, even a nightly routine where my Claude sleeps and reorganizes its memory. That is what taught me how to work around the problems LLMs throw at you, and they come up all the time. Somewhere between four and five thousand hours with LLMs by now, because since I started I never stopped: I failed a lot, and I learned a lot.

The newest of those primitives is measurement. The setup now scores itself on a schedule: whether it recalls what it stored, how much of a session survives the wrap-up, whether the right skill fires, how well each model tier handles a brief it was handed. Closing the first full loop took a day, and more than half of that first round of failures turned out to be problems with the evals rather than with the setup. That is the part nobody warns you about, and it is why I would not trust any of this without the numbers.

Behind all of it is one belief: the bottleneck in enterprise AI isn't the models, it's the translation between them and the people who run critical systems. That's the room I've spent seventeen years in.

Experience

Associate Director, Global Customer Technology Advisor

Kyndryl
2023 – 2026 · Prague

Global Deal Coach for strategically critical mainframe renewals across Europe, APAC, Latin America, and North America.

  • Designed and deployed a 4-agent presales AI system. Cut the business-case cycle from a typical 2 weeks to 1–2 days; 100+ active users globally, escalated by a Kyndryl Fellow to the VP of Data Science, who called it compelling work and flagged its enterprise productization potential.
  • Built a Resource Unit TCO formula from scratch, mathematically decoupling billing complexity from platform utilization. Codified into a £60M 5-year renewal with a major UK insurer, displacing the incumbent competitor.
  • Supported a €180M mainframe cloud renewal with a global European insurer through a year-long negotiation. Addressed dual-CTO concerns with a dedicated session on lessons learned from migration programs delivered by other Kyndryl teams (60+ LPARs with zero backouts; 90+ mainframes consolidated). Established as global technical escalation point.
  • Architected a Mainframe Optimisation Programme: £1.76M implemented savings and a 14.1% MIPS reduction (349 MSUs) across 12 agile sprints.
  • Led ISV software rationalization, challenging legacy licensing assumptions to identify £303K in savings.
  • Architected an EC2-based HA/DR solution for an Italian enterprise (USD 6M contract value, 8,000 COBOL programs, 10M lines of code), mapping mainframe requirements to an AWS target architecture.
  • Reverse-engineered decades of RACF and Adabas security architecture for a government ministry's move to Google Cloud, translating it so Google engineers could implement cloud-native IAM.
  • Created the replatforming architecture curriculum for the Global Modernization Summit program, which ran across six countries (Japan, Poland, Hungary, Czech Republic, Brazil, India). Personally taught the Tokyo and Poland editions, with case studies drawn from engagements led in prior quarters.

Associate Director, Lead Technical Solutions Manager

Kyndryl
2021 – 2023 · Prague
  • Led the first global mainframe-exit workshop for a global European insurer: a multi-city assessment (Milan, Paris, Barcelona, Munich) directing 4–8 Technical Solution Managers. Authored major portions of the 140-page architecture document rated #1 by the customer (60K MIPS environment), with replatforming designed on Micro Focus Enterprise Server on AWS.
  • Built the financial model for a data center hub consolidation, unlocking a 38% revenue increase on future mainframe deals while maintaining client pricing.
  • Led technical defense and financial modeling for competitive engagements, including a mainframe-to-x86 refactoring roadmap reaching positive ROI in year 5 and 176% by year 8.

Senior z/OS System Programmer

Software AG
2020 – 2021 · Darmstadt

Hired to harden mainframe security after a ransomware attack, replacing a retiring sysprog and network architect.

  • Led end-to-end security hardening: AT-TLS encryption across all TCP/IP flows, crypto processor activation, pervasive encryption, and VPN segmentation between LPARs.
  • Migrated 20+ developers from green-screen-only workflows to Zowe CLI and VS Code.
  • Represented the mainframe on the post-incident enterprise security committee during a company-wide security reorganization.

Senior Infrastructure Engineer (z/OS)

Broadcom
2018 – 2020 · Prague

Sole z/OS infrastructure hire in Prague, bridging a 6-hour timezone gap for a 500-developer R&D hub.

  • Eliminated an adversarial R&D-infrastructure dynamic by creating weekly solution interlocks with R&D VPs and Directors, resolving blockers live. The forum dissolved within 18 months as collaboration became self-sustaining.
  • Provided day-0 feedback on 20+ pre-release CA mainframe products, managing complex z/OS beta environments.

z/OS System Programmer & Consultant

IBM
2009 – 2018 · Brazil & Czech Republic

Progressed from mainframe operator through technical lead of a 20-person team to senior systems programmer; later embedded as a consultant at a European financial services client with a 60+ LPAR environment.

  • Automated mainframe operations for a major US insurer, reducing the delivery team from 24 to 12 FTE while maintaining SLA performance.
  • End-to-end z/OS administration across 13 systems: upgrades, hardware migrations, disaster recovery, RACF security, plus IBM pre-GA lab environments.
  • Implemented ITIL and Toyota Lean practices for defect prevention; mentored junior system programmers throughout.

Projects

warmstart

Open-source Claude Code plugin suite for session continuity: context injection, guard hooks, dispatch patterns, wrap-up automation, and an eval suite that scores the plugins' own behavior. Eight public plugins and counting.

github.com/thiagoamaro91/warmstart

Harness evals

Five evals that measure my own development environment instead of a model: memory recall, what survives a session wrap-up, skill routing, dispatch quality per model tier, and action extraction. Runs nightly on two machines, with a regression gate around the nightly memory job.

warmstart: evals chapter

caveat

An agent that reads a Statement of Work from the buyer's side and flags the clauses that quietly shift risk onto the customer.

github.com/thiagoamaro91/caveat

Sessão

A live web product designed, built, shipped, and operated end-to-end with agentic AI development, including automated content pipelines and growth reporting.

sessao.app

Writing

A weekly LinkedIn series about AI adoption from inside a large enterprise: the tooling, the workflows, and the distance between a demo and something you can run every day. Each post ships with working code.

warmstart: session continuity for Claude Code Jul 2026
Context tiering: load what's needed, not everything Jul 2026
Dispatch + skills: making subagents actually work Jul 2026
Workflows: orchestration you can trust Jul 2026
Say hello.

Happy to talk mainframe modernization, AI adoption, or the tools I'm building in public.