How an engagement starts
Most embedded work that goes wrong was scoped before anyone knew what they were looking at. An inherited codebase, a board that has never been powered on, a port to silicon nobody on the team has used — the honest estimate for any of those is a range wide enough to be useless.
So for uncertain work BroadForge leads with a paid technical assessment: a fixed block of hours, typically 12–20 depending on the project, spent finding out what is actually true before either side commits to a schedule.
What the assessment produces
- Review of the available firmware and hardware documentation
- Inspection of repository structure and build system
- MCU, memory, peripheral, and board differences identified
- Toolchain established or verified
- A determination of whether the existing project builds
- Drivers, HALs, middleware, and hardware dependencies mapped
- Assumptions surfaced around clocks, timers, ADC, DMA, interrupts, calibration, memory layout, boot behavior, and FPGA interaction
- Missing equipment, access, files, licenses, or documentation identified
- Major technical risks documented
- A recommended implementation plan
- A supported schedule and cost estimate for the phases that follow
How work is structured
Hourly
When the scope or technical condition is genuinely uncertain, and pretending otherwise would only produce a number nobody should rely on.
Phased
When later work depends on what earlier work discovers. This is most embedded projects.
Fixed fee
Only when deliverables are objectively defined, dependencies are known, acceptance criteria are measurable, client responsibilities are documented, and changes go through a written change-order process.
Rates are quoted against a specific scope rather than published, because a number without a scope attached is not an estimate.