ROLE
Embedded engineering partner across front end, back end, QA, and AI workflows
DURATION
2024 — ongoing
STACK
React · Laravel · AI/automation tooling · AWS
CLIENT
Bizly Inc
WORKFLOW
Agent-augmented engineering · Human-led product calls
ENGAGEMENT
Embedded team, led by Bizly's engineering lead, working directly with internal business stakeholders

WHERE WE STARTED

Bizly didn't come to us because they couldn't ship code. They had an engineering team. The team shipped.

The problem was a different one — the kind a lot of vertical SaaS founders run into a year or two into scaling. The engineers were moving tickets. They weren't thinking like product people. Communication with the rest of the company was thin. The team worked in isolation, not embedded into how Bizly actually operated as a business.

For a SaaS company at Bizly's stage, that gap is expensive. Not in obvious ways — the velocity numbers might still look fine — but in the small daily friction of features that ship slightly wrong, decisions made without product context, and a business team that has to translate everything they need into spec docs because the engineers won't engage outside their tickets.

Bizly was looking for a different kind of engineering partner.

WHAT WE DID

Four PoCs to start

We didn't open with a team proposal. We opened with proofs of concept — four of them, on the front end. Concrete deliverables Bizly could evaluate against the work their existing team was producing.

The PoCs weren't auditions for our engineering ability — they were auditions for the way we work. Could our engineers read a half-formed product brief and ask the right questions? Could they ship something that felt right to use, not just functioned? Could they communicate with non-technical stakeholders without making the conversation harder?

Four PoCs in, the answer was clear. We earned the next layer.

Front-end engineering, with product thinking

We embedded a front-end team into Bizly's existing engineering org. Not as contractors handed tickets at the door — as engineers who joined the standups, the planning conversations, the product discussions. Engineers who pushed back when a brief didn't make sense, surfaced the edge cases the business team hadn't thought of, and shipped features that felt like the company's own work.

This is the core of how we hire. Engineers who can build, but also engineers who can think about what they're building and why it matters to the business.

Back-end work in Laravel

Once the front-end relationship was working, Bizly brought us deeper into the codebase. We took on back-end engineering in Laravel — alongside their existing team, not replacing it.

The goal wasn't a rewrite. It was making the back end faster to work in. Cleaner patterns, better testing discipline, the kind of unglamorous engineering that compounds over time. The work isn't visible to end users, but the team feels the difference every sprint.

Embedding AI into the development workflow

As the engagement grew, we layered AI into how the work itself gets done. Coding agents in the development loop. Automation across the parts of the workflow that don't need a human. AI features built into the product where they actually serve the user — not because AI is on the marketing roadmap, but because specific problems are now better solved with it.

This is how we work across our entire portfolio. With Bizly, the adoption was deliberate and incremental — the team learning the tooling alongside ours, not having it dropped on them.

QA — not just for features, but for empathy

Most agencies' QA tests that the feature works. Ours tests whether the feature feels right to use.

Does the loading state make sense to a user who's seeing this page for the first time? Does the error message tell them what to do next, or does it dump a stack trace? Does the flow respect the user's time, or does it ask for things it doesn't need?

We introduced this kind of QA at Bizly — not as a separate team running tests at the end of the cycle, but as a discipline embedded in how we build. It's product QA, not feature QA. There's a difference, and our clients feel it in their NPS scores.

WHERE IT IS NOW

The Bizly engagement has grown into an embedded engineering team working across front end, back end, QA, and AI workflows.

The structure is the part most agencies couldn't deliver. Our engineers report into Bizly's engineering lead, not into a separate Akvasoft account manager. They join Bizly's product conversations directly — with the business teams, not through a filter. They're treated as part of the company, because for the duration of the engagement, that's what they are.

The work is ongoing. The relationship deepens.

WHAT THIS PROVES

Most engineering teams ship features. Fewer engineering teams think like product people.

The difference matters most when you already have engineers and they're not enough. When the velocity numbers look fine but the work feels off. When the business team is doing too much of the product translation. When you need engineers who join your company's conversation, not engineers who wait for it to be written down.

That's the kind of partner we set out to be. Bizly is the proof that we built it.