10+ engineering teams
- Scope:
- Technical mentoring
Mentoring of engineering teams in languages, cloud and delivery practices.
- PHP, Python and JavaScript/TypeScript
- AWS, test automation and DevOps
Architecture, modernization and mentoring
AI training and technical mentoring for engineering teams
Raise your team's level in AI, Spec-Driven Development, cloud and architecture without stopping delivery, on your real backlog. For CTOs and tech HR.
Free first conversation · Reply within 1 business day
Facts behind this service
10+engineering teams mentored in PHP, Python, JS/TS, AWS, testing and DevOps
The Hackett Group:
Spec-Driven Development with 12 agents, 25 skills, ~50 tasks/week
Regal Rexnord:
Technical leadership of the PHP 5.6 to 8.4 upgrade, zero downtime
Technology leadership of two SaaS products:
Mahout ERP Cloud and Consulta Já
Facts behind this service
10+ engineering teams mentored in PHP, Python, JS/TS, AWS, testing and DevOps
The Hackett Group: Spec-Driven Development with 12 agents, 25 skills, ~50 tasks/week
Regal Rexnord: technical leadership of the PHP 5.6 to 8.4 upgrade, zero downtime
Technology leadership of two SaaS products: Mahout ERP Cloud and Consulta Já
Recognize this situation?
The company rolled out AI coding agents, but nobody trained the team to use them, and quality varies with every pull request.
The company bought online courses, few were finished and almost nothing changed in the way the team works.
Senior developers have no time to coach the others, and the same architecture, cloud and testing questions come back every sprint.
Leadership wants AI in engineering, but there is nobody who has done it in production to show the way.
What we deliver
AI and Spec-Driven Development track
AI and Spec-Driven Development track
The team uses AI to one standard: approved specs before code; Claude Code, Codex, Cursor, Copilot and OpenCode; harness engineering, context engineering, skills, subagents, MCP servers, evals and traces.
Cloud and infrastructure as code track
Cloud and infrastructure as code track
Cloud computing the whole team knows how to run: AWS and GCP with Terraform and Ansible, Docker and Kubernetes, CI/CD, observability, incident response and DevSecOps in the pipeline.
Web development track
Web development track
Web apps that are easier to test and change: React, Next.js, Vue and Angular, PHP/Laravel, Node.js and Python, REST APIs, TDD and end-to-end tests with Playwright.
Desktop and mobile track
Desktop and mobile track
Desktop apps with Electron and mobile apps with React Native and Ionic: app architecture, API and hardware integration, testing, auto-update and app store releases.
Software architecture track
Software architecture track
Tech leads who decide and record the architecture without outside help: DDD and hexagonal architecture, multi-tenant SaaS, event-driven systems and governance with ADRs and RFCs.
Backlog reviews and progress plan
Backlog reviews and progress plan
Reviews of the team's own pull requests, specs and ADRs, with written feedback, and a summary at each cycle with what was practiced, the indicators and the next steps.
See also: AI agent development, AI coding agents, and course content creation.
Results and experience
The Hackett Group (NASDAQ: HCKT)
- Scope:
- Spec-Driven Development with AI coding agents
Engineering operating model with Spec-Driven Development and AI coding agents (Claude Code, OpenCode).
- 12agents and 25 skills, with quality-gate hooks
- ~50tasks delivered per week (406 tasks)
- 39ADRs and 45 specs
Show more results
Regal Rexnord (NYSE: RRX)
- Scope:
- Technical leadership of the PHP modernization
Technical leadership of the upgrade of critical applications, with Claude Code used to reverse-engineer the legacy code.
- PHP 5.6 → 8.4 (10 versions)
- Zero-downtime cutovers
Telecontrol Networking
- Scope:
- Git and Agile adoption
Introduction of Git and Agile methods across the whole company.
- Git and Agile adopted company-wide
- Post-sales, NF-e, finance and reverse-logistics systems
Facing a similar challenge? Tell us the problem and the timeline.
Request a first call
Free first conversation · Reply within 1 business day
Who it is for and who it is not
Technical mentoring is the coaching of an engineering team by senior practitioners who do what they teach, through sessions, pull request reviews and workshops on the team's own code. It covers AI and Spec-Driven Development, cloud and infrastructure as code, web, desktop, mobile, testing and architecture, without stopping delivery.
A good fit for
- CTOs and heads of engineering who want to adopt AI and coding agents with a standard, not each person in their own way.
- Product teams that need to level up in cloud, web, desktop, mobile, testing or architecture without stopping delivery.
- Tech leads, architects and CTOs who want a senior peer to discuss real decisions with.
- Tech HR and leaders looking for AI training for software engineers, applied to the team's own code and tools.
Not a fit for
- Career mentoring, interview preparation or career change into tech: the service is for companies and teams.
- Anyone looking for a recorded course with a certificate of completion: a course platform is a better fit.
- Training beginners in programming: the tracks start from teams that already ship software.
- Talks for large audiences: the format is close work on the team's code.
How it works and timelines
Team discovery
Team discovery
1–2 weeks
We talk to leadership, read the repository and the delivery flow, and map the gaps. You receive a plan with tracks, format, goals and the way progress will be measured, and the proposal.
Sessions on real decisions
Sessions on real decisions
Sessions with the team or with leadership on work in progress: an architecture decision, a spec, an incident. You receive a record of every session, with decisions and next steps.
Reviews on the backlog
Reviews on the backlog
We review pull requests, specs and ADRs from the team's own backlog, with written feedback. You receive the standards that come out of those reviews documented in the team's repository.
Hands-on workshop
Hands-on workshop
When the team needs a leap, we run an intensive workshop on your team's code, for example to adopt AI coding agents. The team leaves with the practice applied to a real task.
Measurement and autonomy
Measurement and autonomy
At each cycle we compare the indicators agreed in discovery with the starting point and adjust the plan. The mentoring ends when the team sustains the practice on its own.
| Scope | Typical timeline | Engagement model |
|---|---|---|
| One-on-one mentoring1:1 sessions with tech leads, architects and CTOs on the team's real decisions | Weekly or biweekly sessions | Monthly retainer |
| Squad mentoringA session with the team plus reviews of PRs, specs and ADRs from the team's own backlog | Weekly session | Monthly retainer |
| Hands-on workshopIntensive work on your team's code, for example to adopt AI coding agents | 1–2 days | Fixed-scope project |
| AI adoption programDiscovery, pilot with one squad, standards, rollout and metrics for the whole engineering team | 4–8 weeks | Fixed-scope project |
| Monthly advisory (office hours)Hours reserved for architecture reviews and leadership decisions | Monthly | Monthly retainer |
Timelines are typical ranges. Discovery defines the scope and the estimate of each project; the contract can be in USD or BRL.
Want to know which format fits your project?
Request a first call
Free first conversation · Reply within 1 business day
Technical details
How we use AI with governance, the decision comparison and the stack
AI with governance
AI is both a subject and a tool of the mentoring. The team learns to use AI coding agents under the rules we use in production: specs before code, quality gates and human review.
Where AI comes inThe team practices with AI coding agents (Claude Code, Codex, Cursor, Copilot, OpenCode) on real backlog tasks, starting from approved specs and your company's AI policies.
Harness and contextThe team learns harness engineering (rules in AGENTS.md and CLAUDE.md, permissions, hooks and checks) and context engineering (memory, compaction, token budget and handoffs between sessions).
Skills and subagentsThe team writes skills with progressive disclosure, triggers, tests and versioning, and uses subagents with isolated context to plan, research and review.
Review, gates and evalsAll agent-generated code goes through a person on the team and through test, coverage and static-analysis gates in CI; evals and traces (OpenTelemetry, LangSmith, Langfuse) measure the agent.
Autonomous AI agentsOpenClaw and Hermes Agent (Nous Research), with permissions, channels, scheduled tasks and persistent memory under control, once the team masters the harness.
Security and dataThe team learns to handle prompt injection, output sanitization and tool permissions before connecting agents to internal systems; we work under NDA, and client data never trains models.
Decision comparison
| Criterion | Online course | In-house workshop with an instructor | Mentoring on the team's backlog |
|---|---|---|---|
| Content | Generic, the same for every company | Adapted to the topic, with prepared exercises | The team's own code, tools and decisions |
| Who leads | A recorded instructor, with no contact with the team | An instructor who specializes in teaching the topic | Senior engineering that ships production systems |
| Application at work | Left to each person, after the course | Left to the team, after the event | Happens during the mentoring, on backlog tasks |
| Follow-up | None, or a Q&A forum | Ends with the event | PR and spec reviews over the following weeks |
| When it makes sense | Fundamentals and individual study, at each person's pace | Introducing a new topic to many people at once | Changing how the team works, with results in the code |
Stack and integrations
- Claude Code
- Codex
- Cursor
- Copilot
- MCP
- AWS
- GCP
- Terraform
- Ansible
- Docker
- Kubernetes
- GitHub Actions
- React
- Next.js
- Laravel
- Electron
- React Native
- Playwright
Frequently asked questions about Technical Mentoring & AI Training
Who is the technical mentoring for?
For companies with an engineering team: CTOs, heads of engineering, tech leads and product squads that already ship software and want to level up in AI, cloud, web, desktop, mobile, testing or architecture. It is not career mentoring or a course for beginners. The format changes with the audience, from one person to a whole engineering team: one-on-one mentoring for leaders, squad mentoring for product teams and an AI adoption program for several squads, starting with a pilot.
Which technologies does the technical mentoring cover?
The mentoring covers AI and Spec-Driven Development (SDD), cloud computing and infrastructure as code (IaC), web, desktop and mobile development, testing, DevOps and software architecture. In practice: coding agents such as Claude Code, Codex, Cursor and Copilot; AWS and GCP with Terraform, Ansible, Docker and Kubernetes; React, Next.js, Vue, Angular, PHP/Laravel, Node.js and Python on the web; Electron on the desktop; React Native and Ionic on mobile. Discovery picks the tracks from what your team uses today and what needs to change.
Is the mentoring remote or in person?
The mentoring is remote: sessions by video call and reviews directly in the team's repository and pull request tools. Master Timm works from São Paulo, Brazil, with companies in the United States and Brazil. São Paulo is 1–2 hours ahead of US Eastern Time, so working hours overlap. An in-person workshop is evaluated case by case, depending on location and schedule.
How is the team's progress measured?
By indicators agreed in discovery and taken from the engineering flow itself: test coverage, pull request review time, deployment frequency and adoption of the agreed standards. At each cycle you receive a summary of what was practiced, what changed in the indicators and the next steps. The final measure is the team sustaining the practice without the mentoring.
Does the mentoring use our own code and tools?
Yes. The mentoring happens in the repository, the backlog and the tools your team already uses, not in generic exercises. We review pull requests, specs and ADRs from work in progress, under NDA, and follow the AI and security policies of your company. If the team has no approved coding agents yet, choosing the tools is part of discovery, together with your security team.
Is the mentoring available in English and in Portuguese?
Yes, the mentoring is delivered in English or in Portuguese, and we have worked with US companies since 2022. Sessions, reviews and materials follow the team's language. In teams spread across countries, the default is English, including the written feedback on pull requests and specs.
Do you offer AI training for software developers and engineering leaders?
Yes. For developers it is hands-on, on the team's code, and includes Claude Code training for developers: Spec-Driven Development with Claude Code, Codex, Cursor, Copilot and OpenCode, harness engineering, context engineering, skills, subagents, MCP servers, LLM engineering, RAG, guardrails, evals and traces. Autonomous AI agents, such as OpenClaw and Hermes Agent (Nous Research), come after that foundation. For leaders, the focus is decisions: where to adopt, how to measure, which risks to control and how to organize the rollout. Both audiences meet in the 4–8 week AI adoption program.
How much does technical mentoring for a team cost?
The cost depends on the format, the frequency of the sessions, the number of people and the tracks chosen. That is why we do not publish a price list. The first conversation is free; if it makes sense to proceed, a 1–2 week team discovery defines tracks, format and goals, and the proposal follows it. Mentoring and advisory are hired under a monthly retainer; workshops and the AI adoption program as fixed-scope projects, in USD or BRL.
About Master Timm
Master Timm
AI-focused IT consultancy
Senior engineering from discovery to delivery: whoever designs the architecture also writes and reviews the code.
- 20+ years of software engineering experience
- Projects for US and Brazilian companies, under B2B contracts
- Specs, tests and documentation in the client's repository
- Fixed scope, capped time & materials or retainer, in USD or BRL
Why Brazil
- São Paulo is 1–2 hours ahead of US Eastern Time, so working hours overlap.
- We work in English: calls, specs and documentation.
- Remote B2B work for US companies since 2022.
- Scope, ownership and confidentiality are set in the contract before work starts.
Master Timm · Updated on
Let's plan your team's track
In the first conversation we look at where the team stands, leadership's goals and the backlog, and suggest the tracks and the format that make sense. You talk directly with whoever will lead the mentoring.
Request a first call
Free first conversation · Reply within 1 business day