The suite

One platform for the whole
FPGA-to-ASIC correctness flow

A family of focused engines and autonomous pilots - starting with review and verification today, extending across implementation, place-and-route and timing closure.

Product family

One workspace, organized by maturity

Engines do one job extremely well; pilots orchestrate engines toward an outcome - with the engineer in the loop. Here's what you can use today, what's actively being built, and what's on the roadmap.

Available today

Beta

ReviewForge engine Beta

Static analysis plus an engineering reviewer over your existing RTL - bugs, synthesizability, CDC and reset risks, with generated assertions. The lowest-risk way to put AI into your flow.

VeriForge engine Beta

Self-checking functional verification: directed and constrained-random testbenches, assertions, coverage, and a reference-model scoreboard that scores itself.

Playground workspace Beta

The live workspace to run ReviewForge, VeriForge and the full loop in your browser - pick a target, give a spec, and watch each stage stream.

In active development

Experimental

RTLPilot pilot Experimental

The engineer-in-the-loop flow from spec to verified RTL - create, verify, repair and report PPA across FPGA vendors and ASIC technologies, with you approving before anything ships. Runs end-to-end today; maturing fast.

Future platform

Coming soon

FPGAPilot pilot Coming soon

RTL → FPGA bitstream, driving implementation with timing-aware optimization iterations.

TimingForge engine Coming soon

Static timing analysis and closure across the implementation flow.

PnRForge engine Coming soon

ASIC place-and-route engine for the physical-design stage of the flow.

ASICPilot pilot Coming soon

Full RTL → GDSII orchestration - the long-term vision the rest of the suite builds toward.

Roadmap shown for direction; availability and sequencing may evolve.

ReviewForge

Review, the trust-building entry point

Static checks plus an engineering reviewer that reads your design for functional and synthesizability risks and writes assertions to lock down invariants. Pure critique - it never rewrites silicon you didn't ask it to.

  • Categorized findings with severity & cited signals
  • Suggested assertions you can paste straight in
  • Onboard legacy IP, or a second pair of eyes on a tricky FSM
Open ReviewForge →
// ReviewForge · 3 findings · 2 assertions HIGH latch inferred on `next_state` MED CDC on `busy` - add synchronizer LOW width mismatch in `cnt` assignment ASSERT !(full & empty) ASSERT wr_en & full |-> ##1 $stable(count)
VeriForge

Verification that scores itself

Give VeriForge a design and a spec; it builds a testbench with directed corners, a constrained-random phase, assertions and a reference model - then runs it and reports a pass/fail score and coverage. No manual waveform review.

  • Pass/fail per check + coverage summary
  • Waveform captured for every run
  • Sandboxed, time-bounded simulation
Open VeriForge →
sync_fifo · functional waveform34/34 checks ✓
clk wr_en full dout[7:0]
Trust & security

Your RTL is your crown jewel

Nexeda is designed to run where your IP already lives. A self-hosted / VPC deployment is a first-class path for design partners, not an afterthought.

On-prem / VPC

Run the full loop inside your own environment - no RTL leaves your network.

Reproducible engine

Built on a hardened open simulation & synthesis backend, so results are inspectable and repeatable.

Every artifact saved

RTL versions, testbenches, logs, waveforms and PPA reports - auditable at each iteration.

See it run on your kind of block

Try ReviewForge and VeriForge in the live Playground.

Open Playground How the loop works