Bar Zik

Approach

How the work actually runs.

No discovery theatre, no forty-slide readouts. Five phases, written deliverables, and a standing weekly session where decisions get made with your team in the room.


  1. Conversation

    45 minutes, free

    You describe what you are building and where it is stuck. I ask a lot of questions. By the end of it we both know whether there is an engagement here, and I will tell you plainly if there is not.

    Output A clear yes or no, and if no, a pointer to something more useful.

  2. Scope

    2–4 days

    I write a short proposal: the specific question the engagement answers, what I will deliver, the timeline, and a fixed price. If the scope is wrong, we fix it before anyone commits.

    Output A one-page proposal with a fixed price and dated deliverables.

  3. Immersion

    First week

    I get into the codebase, run the system against real inputs, read your traces, and talk to the engineers who built it. This is where most of the actual findings come from — not from the architecture diagram, but from watching the thing behave.

    Output Shared working notes, updated as I go. No black boxes.

  4. Work

    Core of engagement

    Design, build, or both, depending on the engagement. We run a standing weekly session so decisions get made with your team rather than handed to them. Everything gets written down with the reasoning attached.

    Output Documented decisions, working code where relevant, weekly sessions.

  5. Handover

    Final week

    A walkthrough with the engineers who will own this, plus written documentation that assumes I will not be around to answer questions. If the team cannot run with it independently, the engagement has not finished.

    Output Documentation, walkthrough session, and a prioritised what-next list.

Working principles

Commitments that hold on every engagement.

I write things down.

Every finding, decision, and recommendation arrives as written prose with the reasoning included. Slide decks compress out exactly the parts your engineers need six months later.

I work in your codebase.

Reviews based on architecture diagrams and conversations miss the things that actually break. I read the code, run the system, and look at production traces.

I price the engagement, not the hours.

Fixed scope, fixed price, agreed up front. Hourly billing rewards slow work and makes you hesitate before asking a question.

I stay independent.

No reseller agreements, no vendor partnerships, no referral fees. When I recommend a model, database, or framework, that recommendation is the whole reason.

I say no.

If your problem is not one I can solve well, I will tell you in the first conversation. A bad-fit engagement costs you a quarter and costs me a reputation.

I aim to become unnecessary.

The goal of every engagement is a team that can make the next decision without me. Retainers exist for teams who want ongoing review, not ongoing dependency.

Next step

The first conversation costs nothing.

Forty-five minutes, no proposal attached. You will leave with a clearer read on your problem whether or not we work together.