About · Ascend Griffin

Most IT projects don't fail because of bad code. They fail because no one on the client's side had the technical fluency to see it coming.

I'm Taras Smaliukh — a buyer-side IT ally for non-technical founders and vibe coders. I act exclusively in your interests — from the first conversation about your idea to a working outcome delivered.

↓ Expand a section to read more

Why you can trust me

Acting CEO of TreximAI · 8 years BA across AI products, logistics, fintech, B2B SaaS

23+
IT projects shipped
Engagement formats
Audit · Discovery · Project Support
Based in
Lviv, Ukraine · fully remote · travel when it helps

Cases on this site are illustrative patterns drawn from 8 years inside IT — not invented for the page. No published clients yet.

01 Who I am

I've spent 8 years inside IT as a business analyst and product owner — across AI products, logistics, fintech, and B2B SaaS. I've sat on both sides of the table: negotiating with vendors, owning product backlogs, running discovery, building teams from scratch.

My original training was in safety-critical systems — Air Traffic Control, where unclear requirements don't produce bad UX, they produce accidents. That discipline carries into everything I do with products now.

I'm currently CEO and Chief Product Officer at TreximAI, a B2B AI logistics platform — banking, fuel retail, insurance, e-signature. I built the team and the product from zero, secured six-figure development capital, and negotiated agreements with incumbents across regulated sectors.

That experience isn't a line on a CV — it's the reason I understand how IT projects actually behave under pressure. Not from books and not from slides — from practice, where the cost of being wrong is real.

02 What I do

Consultants advise. I take ownership of the outcome.

Concretely: I read vendor contracts and tell you which clauses will cost you in month three. I run structured discovery so you're not building the wrong thing. I sit on demos and ask the technical questions you don't yet know to ask. I review estimates and spot the assumptions that turn a $40K project into a $70K one.

When something goes wrong, I'm the person who says "here's exactly what broke and here's the fix order" — not "let me introduce you to another consultant."

I'm not a technical architect. I won't write your code. But I will make sure the people writing it are accountable to your interests, not theirs.

03 How I work

Three ways we work together.

Discovery comes first, always. A 30-minute conversation to understand your situation. If I'm not the right fit, I'll say so — and usually recommend who is.

  1. A

    Solo advisory.

    I work directly with you: reviewing deliverables, joining key calls, providing structured oversight of your vendor or in-house team. You keep control of your team; I'm the experienced voice in the room on your side.

  2. B

    Full delivery with my network.

    For founders who want one accountable contract — one person responsible from brief to working product. I coordinate a tight, tested network of engineers, designers, and QA specialists. One point of contact, one contract, clear milestones with fixed pricing.

  3. C

    Discovery and handoff.

    We validate what you're building, produce a complete product brief and scoped requirements, and I help you find and brief the right vendor. You own the result. I help you start right, not stay forever.

Pricing is fixed and agreed before work starts. No hourly billing.

04 8 years of practice + AI tooling

Why this is a different level of work than five years ago.

Ascend Griffin launched in April 2026. It's a new form for the work, but not new work — 8 years of business analysis and product ownership are underneath. What changed is the tooling and the engagement model.

AI tools changed the depth of what's possible. Reading a 50-page vendor proposal — an hour now, not two days. Discovery with a client — deeper, because the AI agent keeps the full context. Documentation — generated, I just validate. The same instinct from 8 years of practice applies significantly deeper, faster, and at lower cost than it used to.

A different engagement model. I work directly with the person who actually has a stake in the outcome — not with hours, not with process, not with internal agency politics. That's a different contract than what a typical IT-partner relationship offers.

The illustrative patterns below describe the kinds of situations I work on — real problem types, not invented for the page. They're honest about what they are: composites, not client testimonials.

05 Who I am for

Four kinds of founders I read for.

A

The vibe coder who built something real.

Illustrative — a pattern I work with often.

A founder built a working prototype in Cursor. Twenty paying users, a few features to add, and a growing feeling that something in the foundation isn't quite right. They're not a developer — they don't know how to read the codebase, only how to prompt into it. What they need is someone who can look at what they've built and tell them: what holds, what doesn't, and what the path to production-readiness actually looks like.

B

The non-technical founder about to sign a contract.

Illustrative — the situation that comes up most often.

A business owner has three vendor proposals. The numbers range from $45K to $120K for what sounds like the same project. One vendor is confident, two are vague, all three are professionally persuasive. Without a technical background, it's nearly impossible to tell which quote is honest, which is padded, and which is going to grow by 40% the moment a change request lands.

C

The founder with an idea and domain expertise.

Illustrative — an early-stage pattern.

Someone with deep knowledge of a specific industry — logistics, fintech, healthcare admin — has been thinking about a software product for two years. They have domain expertise most developers will never have. What they don't have is a clear answer to: should this be built at all? What would it actually cost? What would need to be true for it to succeed?

D

The founder whose product is quietly breaking.

Illustrative — a rescue pattern.

A team shipped a product. It works, mostly. But delivery velocity has slowed to almost nothing, the original vendor is less responsive, the budget is burning on maintenance without visible progress, and no one can clearly explain why. The first step isn't re-architecting. It's a clear-eyed diagnostic: what broke, when, and why.

06 When I recommend you look elsewhere

Some questions aren't my strongest read.

A short, honest map of when a different conversation will serve you better:

  • Pitch deck or fundraising help. A separate craft with its own specialists. I can point you to people who do this better than I would.
  • Recruiting your team. I'm not a headhunter. I can help you scope who you're looking for and how to interview them — but the actual hiring process belongs with others.
  • Pure technical architecture without a business angle. If the question is only about tech stack, there are architects who'll go deeper than I would.

Next step

Discovery Call — 30 minutes.

Book Discovery Call →

GET IN TOUCH

Let's talk

I'll get back to you within 24 hours — to schedule a discovery call or discuss your inquiry.

Prefer direct contact?

or send a message

Or email directly: taras@ascendgriffin.org

Got it. Talk soon.
I'll be in touch within 24 hours.