Frequently asked questions

Four primary models: a founder-friendly devshop for end-to-end product builds, AI applications development for AI features and copilots, AI-First engineer supply for AI-native talent, and staff augmentation (outstaffing) for teams that need senior engineers embedded into their squad.

For dedicated pods and staff augmentation we charge a transparent monthly seat rate per engineer. For end-to-end product builds we offer a fixed-cost MVP option after a paid two-week scoping sprint. AI application engagements are scoped per pilot, then move to a monthly run-and-improve retainer.

Yes. Many clients start with a fixed-cost MVP and graduate to a dedicated pod once the product is in market. Others start as staff augmentation, then take on a fixed-scope feature as a separate engagement.

Yes, completely. IP assignment is part of every MSA. You own the source code, the design assets, the model weights you train, and the production infrastructure from commit one.

Conversational copilots, retrieval-augmented generation (RAG) systems over private data, document automation, AI-powered search, internal AI workflow tools, and agentic systems with human-in-the-loop checkpoints.

No. We are LLM-agnostic. We pick OpenAI, Anthropic, Google, or open-source models on a per use-case basis, based on quality benchmarks, latency, and cost. We also build evaluation suites so model swaps stay safe.

Every AI project starts with a threat model and a data-handling brief. Production builds include PII redaction, structured audit logging, evaluation gates in CI, and human-in-the-loop approval for irreversible actions.

Yes. We integrate via API, event streams, or webhook patterns into existing CRMs, ERPs, knowledge bases, ticketing systems, and data warehouses. AI features are delivered as a service inside your stack, not as a separate product.

Every Kiebot engineer is trained on vector databases, LLM orchestration frameworks, eval systems, and AI-assisted coding tools. They think probabilistically when they design systems, and they pair-program with AI to ship faster without lowering the quality bar.

Engineers go through a multi-stage process: resume screen, async coding task, live system-design interview, AI-fluency conversation, and a final cultural fit interview. Only ~5% of applicants reach our active bench.

Median experience on our bench is six years. We deliberately keep our pods senior-heavy because senior engineers ship cleaner code and need less hand-holding from your team.

Three non-negotiables: every PR is reviewed by a senior; CI runs the same gates as production; and we treat tests, observability, and runbooks as part of the deliverable, not an afterthought.

Two-week sprints with a planning meeting on Monday, daily 15-minute async standups, a working demo on Friday, and a retrospective at the end of the sprint. Your product owner approves what enters the sprint.

We guarantee a minimum four hours of working-day overlap, calibrated to your business hours. For European, UK, and US clients this is built into the contract from day one.

Yes. Whether you engage us as a devshop or as staff augmentation, our engineers work inside your tooling stack. We do not run parallel ticketing systems unless you specifically ask us to.

Every scope change becomes a written impact assessment with the new ask, the trade-off, and the recommended next step. Sponsors choose, but they always choose with data.

Yes — mutual NDAs, DPAs, and BAAs (for healthcare) are routine. We can also sign region-specific agreements such as UK ICO data-processing addenda or HIPAA business-associate agreements.

Kiebot does not store client data unless explicitly required by the engagement. Engineers work inside your cloud accounts and your tooling. Where shared workspaces are required (e.g. shared design files), we use your tooling preferences.

Yes. We have shipped products inside GDPR, HIPAA, and PCI-DSS scope. Each engagement begins with a compliance brief so engineers join with the right awareness from sprint one.

You can ask for a replacement at any time. We typically place a replacement within two weeks, with a documented handover from the previous engineer.