Powdrix is an agent operating system for production 3D-printing floors. An as-built part twin picks the recipe before the build, edge agents steady the melt pool layer by layer and control recoat, powder and thermal distortion, CT and vision grade every part against qualification criteria, robots handle plates, powder and post-processing, and a floor brain predicts build failure and schedules the fleet, all bounded by a policy engine and the qualified parameter envelope you control.
Illustrative scenario: PBF-2 · Ti-6Al-4V bracket family, 24-part build, 23 of 24 parts released, one held
Design partners wanted · 3 to 5 floors · machines and cells we are looking for
Pre-launch. No customers, pilots or logos are claimed. The first cohort gets a paid pilot on one machine against a baseline agreed in writing.
Every Powdrix run is an ordered, inspectable sequence. This is run_7c2e4b on PBF-2 · Ti-6Al-4V bracket family, 24-part build. Scenario completes in 10 steps with one approval gate. Twin recipe #14 selected, one lack-of-fusion precursor at layer 1,872 cleared by a bounded power trim, CT triage in 11 minutes, 23 of 24 parts released with full genealogy and one held for review.
Floor Orchestrator · Pulled the job from AM MES: 24 Ti-6Al-4V brackets, 60 µm layers, 3,140 layers, density target 99.7%, geometry tolerance ±0.10 mm, AS9100 evidence packet required. Locked the OEM-approved parameter envelope for PBF-2.
Part Twin · Simulated 40 candidate orientation, support and parameter recipes for thermal field, residual stress and distortion, ranked on predicted density, geometry deviation, takt and energy.
Prep-and-Parameter · Recipe #14 released: 32° orientation, lattice supports on two faces, laser power and speed inside the qualified envelope. Predicted distortion 0.06 mm, predicted density 99.8%.
Recoat-and-Powder · Recoater image checked every layer. Powder lot L-2291 on its third reuse, humidity in window. One short-feed at layer 412 corrected with a dosing increase before the next layer fused.
Melt-and-Layer · Melt-pool, photodiode and thermal streams fused at 120 fps. Lack-of-fusion risk rose to 0.61 on part 17 at layer 1,872 and was cleared by a 3% power trim inside the envelope within four layers.
Thermal-and-Distortion · Interlayer temperature on the tall brackets ran 14 °C above the twin's prediction from layer 2,300. A 1.5 s interlayer dwell held the thermal field inside the predicted band through the top section.
Floor Orchestrator · Melt-and-Layer proposed a hatch-spacing change on part 17 beyond the qualified envelope. Held for the AM process engineer, who declined it and chose to flag part 17 for CT priority instead. Decision logged.
Robot-and-Handling · Plate unloaded by the handling cell on build complete, depowdered, and parts transferred to stress relief. Powder returned to the sieve with lot L-2291 reuse count incremented in genealogy.
Defect-and-Inspect · 24 CT volumes segmented for pores, cracks and lack of fusion in 11 minutes. 23 parts graded accept against the acceptance criteria. Part 17 shows a 180 µm indication at layer 1,870 and is routed to the inspector for disposition.
Quality-and-Conformance · 23 parts released with the full genealogy: recipe, layer tags, melt-pool events, thermal trace, approvals, post-processing cycles and CT result. Part 17 held with a non-conformance opened. Evidence packet assembled for sign-off.
Every tool invocation, argument and result is written to an append-only, human-readable log, and every reasoning step is expandable. Nothing about a build is hidden from the floor or from the certification engineer.
Plan, thought, action, observation, exactly as the orchestrator would record it.
Run job J-4471, 24 Ti-6Al-4V brackets on PBF-2, to 99.7% density and ±0.10 mm, with an AS9100 evidence packet on every released part.
Simulate the recipe on the twin first. Release only inside the qualified envelope. Watch recoat and melt pool every layer, compensate thermally against the twin, and route anything outside the envelope to the process engineer. Triage every part by CT before release.
Three candidate recipes beat #14 on takt but predicted distortion above 0.08 mm on the tall brackets. Takt is not worth a geometry failure on a qualified part family. Recipe #14 wins on the stated targets.
twin.simulate(candidates=40) returns recipe #14: 32° orientation, lattice supports, power and speed inside the envelope, predicted distortion 0.06 mm.
Lack-of-fusion risk 0.61 on part 17 at layer 1,872. A 3% power trim, inside the envelope, clears it in four layers. The proposed hatch change would leave the envelope, so it goes to a human and is declined.
Flag part 17 for CT priority. Hold interlayer dwell at +1.5 s from layer 2,300 to keep the thermal field inside the twin's band.
CT confirms a 180 µm indication on part 17 at layer 1,870, two layers below the precursor. The other 23 parts grade accept. The precursor, the action and the CT result are now one linked record.
Scenario complete. 23 parts released with full genealogy, part 17 held with a non-conformance, CT triage in 11 minutes, no unapproved parameter left the envelope.
Nothing about a Powdrix build is implicit. The job is declared, the recipe is simulated on the twin, each step is typed and bounded by the qualified envelope, and the outcome is written to the build genealogy.
A run starts as a job, becomes a simulated recipe, and executes as a graph of typed steps across the printer, the ovens, the robot cell and the CT. The active path is always visible, always logged, always reversible.
L1 advisory, L2 supervised writes, L3 bounded autonomy inside the OEM-approved envelope, L4 unattended within an envelope once validated. Each floor sets the level per agent, per machine, per part family, and anything that affects qualification waits for a named approver. The four figures above are design targets, not measured results.
Each agent owns a stage of the additive floor, a bounded tool set and a metric the floor already measures. They negotiate for shared machines and cells through the orchestrator, never directly. Five of them ship as named products.
Build preparation, orientation, supports, nesting, process parameters
Proposes orientation, support, nesting and process-parameter recipes (laser power, speed, hatch, layer) for a build, scores them against density, geometry and takt targets with the part twin, and releases the winning recipe to the printer only inside the OEM-approved parameter envelope and after sign-off.
In-situ melt-pool, layer and energy control
Fuses coaxial and off-axis melt-pool video, layer and recoater images, photodiode and thermal streams at high frame rate, classifies and predicts porosity, lack of fusion, keyholing, spatter and recoat faults seconds ahead, and closes into bounded, operator-graduated laser and energy control to hold the pool stable.
Recoating, powder-bed uniformity, powder quality and reuse
Watches recoater images and bed uniformity every layer, detects streaks, short feeds and spreadability faults before the next layer is fused, tracks powder lot, reuse count and quality, and recommends or applies bounded recoat and dosing corrections.
Thermal fields, residual stress, distortion
Tracks the thermal history of the build from in-situ thermal streams against the twin's prediction, estimates residual stress and distortion as the part grows, and recommends thermal compensation, scan-strategy or dwell changes before geometry leaves tolerance.
CT, vision and metrology inspection, grading, qualification evidence
Triages CT volumes for pores, cracks and lack of fusion with GPU 2D and 3D segmentation, fuses surface vision and metrology deviation, grades each part against the alloy and part acceptance criteria with a human sign-off checkpoint, and links every defect back to the in-situ signal and the recipe that produced it.
Stress relief, heat treatment, HIP, support removal
Holds stress-relief, heat-treatment and HIP recipes against the part's as-built state and the qualification spec, sequences support removal, and records each thermal cycle into the part genealogy so post-processing stops being the blind, reactive end of the floor.
Yield, build-failure risk, energy, fleet scheduling
Ingests build health, inspection results, recipes and cell state from the whole floor, predicts build-failure and qualification risk per job, and schedules machines, ovens, HIP, inspection cells and robots for yield, takt and energy with cuOpt, re-optimising after a delay or a quality hold.
Build-plate, powder, part handling and depowdering
Drives Isaac and Jetson robot cells that load and unload build plates, handle and sieve reactive powder, depowder parts and move them through heat treatment, HIP and support removal, with motion planned and checked in simulation and coordinated with live printer and oven state.
Right-first-time, non-conformance, build genealogy, qualification traceability
Maintains the immutable build genealogy from recipe to layer to inspection to disposition, runs non-conformance workflows, and blocks shipment when genealogy or defect evidence is incomplete, so every qualified part carries a record a certification engineer can audit.
Parameter sets, specs, AM standards, procedures, operator Q&A
Answers process and qualification questions from parameter sets, build and qualification specs, AM standards and quality procedures, with every answer cited to its source and scoped to the tenant, so engineers get grounded reasoning rather than a chat window.
As-built part simulation and recipe optimisation before the build
Simulates the build for candidate orientation, support, nesting and parameter recipes: thermal fields, a melt-pool proxy, residual stress and distortion. Auto-optimises against density, geometry, takt and energy targets, stress-tests the winner against rare defect and distortion scenarios, and hands the validated recipe to Meltrix.
Planning, arbitration, human approval
Plans the build job, routes events to the right agent, arbitrates between agents competing for the same machine or cell, enforces the autonomy level and the OEM parameter envelope, and sends any qualification-impacting action to a named approver.
Powdrix is not a dashboard. Each agent closes a loop an additive floor already measures in scrapped builds, failed qualifications, CT hours and idle machine time. The tables show what each loop watches and what it is allowed to move.
Melt-pool video, photodiode and thermal streams are fused at the printer edge into a rolling risk score for porosity, lack of fusion and keyholing. When risk crosses the threshold the agent has layers, not hours, to trim power inside the envelope instead of scrapping the build at CT.
| Signal | Watches | Moves |
|---|---|---|
| Melt-pool signature | Watched per layer, per part | Power and speed trim inside the envelope |
| Keyhole and spatter | Classified from video and photodiode | Focus and speed recommendation |
| Lack-of-fusion risk | Scored continuously | Trim, CT priority flag, or hold request |
A recoat fault at layer 400 ruins a build that finishes a day later. Recoat-and-Powder reads the recoater image every layer, tracks lot, reuse and humidity, and corrects dosing or asks for a hold before the next layer is fused.
| Signal | Watches | Moves |
|---|---|---|
| Bed uniformity | Recoater camera, every layer | Dosing and recoat speed inside limits |
| Powder quality | Lot, reuse count, humidity | Reuse limit and sieve request |
| Streaks | Detected before fusion | Hold request to the orchestrator |
Spot CT on a sample and hours of manual review cannot keep pace with serial production. Poroscan segments every volume on GPUs, grades against the acceptance criteria with a human sign-off, and links the defect back to the layer and the signal that produced it.
| Signal | Watches | Moves |
|---|---|---|
| Porosity and cracks | CT volume segmentation | Accept, reject or inspector routing |
| Surface faults | Vision inspection | Rework or reject suggestion |
| Geometry | Metrology deviation | Compensation feedback to Parttwin |
Every recipe is simulated for thermal field, residual stress and distortion before release. Candidates that save takt but push distortion or predicted density outside target are discarded automatically, and the winner is stress-tested against rare defect scenarios.
| Signal | Watches | Moves |
|---|---|---|
| Orientation and supports | Thermal and stress simulation | Validated recipe to Prep-and-Parameter |
| Parameters | Melt-pool proxy and density prediction | Envelope-checked recipe to Meltrix |
| Distortion | As-built versus as-designed | Compensation and predicted-defect map to Poroscan |
Two agents will want the same machine or the same parameter. The orchestrator arbitrates on the job goal and the qualification rules, not on who asked first, and the handoff is logged like any other step.
Yield-and-Takt wants to raise scan speed on the build to recover takt lost to the recoat correction at layer 412.
Melt-and-Layer wants to trim laser power on part 17 to clear a lack-of-fusion precursor at layer 1,872.
Lack-of-fusion risk 0.61 on a qualified part outranks 1.1 hours of takt under the job goal. Melt-and-Layer wins the parameter for four layers; the speed change is deferred.
Yield-and-Takt recovers the takt by re-sequencing the oven queue with cuOpt instead. Density prediction holds at 99.8%.
Illustrative scenario.
A qualified floor does not go from manual to unattended in one step. Powdrix makes the level explicit, auditable and reversible at any time, and human approval for anything that affects qualification stays mandatory until validation exists. The first release runs shadow mode, then assist mode, then graduated autonomy.
| Level | What the agent does | What the human does | Where it sits in a pilot |
|---|---|---|---|
| L1 · Advisory | Predicts and recommends, with citations; nothing is written | Enters every change manually | Shadow mode on one machine, first weeks of a pilot |
| L2 · Supervised | Proposes a parameter, thermal or inspection move; it executes on approval | Approves each move in the review console | Assist mode once accuracy is measured against CT |
| L3 · Bounded | Writes inside the OEM-approved parameter envelope and per-tag limits | Approves anything that leaves the envelope or affects qualification | Graduated autonomy on low-risk loops, if the floor is satisfied |
| L4 · Unattended | Runs the envelope without prompting | Sets the envelope; reviews the shift record | Planned; only after validation on the part family |
These are the figures in our compute plan. None is a measured result. Every pilot measures against the floor's own baseline, reproduced from the floor's build records and CT outcomes, not from our telemetry.
Aerospace, medical and dental implants, energy and defence, tooling and industrial, and service bureaus, on metal powder-bed fusion, directed energy deposition and polymer SLS, MJF and FDM. Most floors pick one machine and one wedge, prove the number, then expand.
Qualified part families where one failed build or failed qualification is expensive. Meltrix and Poroscan on one metal PBF printer first, with the AS9100 evidence packet attached.
Porosity and surface faults are patient safety. CT grading against acceptance criteria with a named sign-off, genealogy for every implant.
Serial production under qualification and reshoring pressure. Build-failure prediction and fleet scheduling with Yieldex as the floor scales.
Cost per part decides the business. Parttwin to cut coupon campaigns and ramp scrap, Forgebot to reclaim idle machine time.
Many customers, many part families, one floor. Yieldex scheduling across machines, ovens and inspection, with per-customer genealogy.
A build is thousands of layers and a day of machine time. Perception has to be local, deterministic and fast, CT has to be triaged in minutes, and the twin has to run before the build. Powdrix runs GPU inference at the printer edge and keeps training, simulation and optimisation in the cloud or on-prem.
One Jetson node per printer for 30 to 240 fps melt-pool, layer and recoater streams with photodiode and thermal fusion. Sub-250 ms alerting and 1 to 2 s bounded recommendations are the design targets. Jetson Thor is planned for multi-laser machines.
2D and 3D segmentation models for porosity, cracks, lack of fusion, surface defects and metrology comparison, served per inspection cell. Sub-15-minute CT triage per build is the design target.
Defect, melt-pool-instability, build-failure and AM reasoning models trained weekly over multimodal build data, with gated promotion from shadow to assist to bounded. Dedicated capacity is planned.
Thermal, melt-pool-proxy, residual-stress and distortion simulation for recipe ranking, plus robot-cell and queue simulation. Modulus surrogates are planned so sweeps run at production cadence.
Rare pore, lack-of-fusion, spatter, recoat-streak and distortion scenarios generated and validated against real CT and in-situ data before promotion. 1,000 to 10,000 per qualified part family is the initial target.
Build nesting, fleet scheduling across machines, ovens, HIP, inspection and robots, energy windows, and GPU analytics over genealogy tables.
Plate, powder and part handling policies trained in simulation and run on Jetson in the cell. NIM packages perception, prediction and reasoning services behind the floor orchestrator.
The Powdrix SDK is typed Python. Tools are declared with schemas, a qualified parameter envelope and limits, and the policy engine enforces them at call time, not in a review meeting. This is a design preview; the API will change before release.
# Design preview. Bound the melt-pool agent to printer PBF-2.
from powdrix import Agent, Tool, Envelope, Autonomy
power = Tool(
name="printer.trim_power",
tags=["PBF2.LASER_POWER", "PBF2.SCAN_SPEED"],
envelope=Envelope(qualified="BRK-TI-A", max_step="3%", per="4 layers"),
)
melt = Agent(
id="agent.melt_layer",
goal="density >= 99.7%, lack_of_fusion_risk < 0.2",
tools=[power, Tool("holoscan.fuse", read_only=True)],
# bounded writes inside the envelope; humans gate anything outside it
autonomy=Autonomy.L3,
# check the move against the twin before it is written
verify="twin",
)
run = melt.start(printer="PBF-2", job="J-4471")
for step in run.stream():
print(step.layer, step.name, step.status)The same run engine, the same policy checks, the same audit trail, from the terminal, the review console or the SDK. Design preview.
$ powdrix run "build J-4471 on PBF-2" --autonomy L3
→ plan composed 10 steps · 1 approval gate
→ twin.simulate 40 candidates · best #14 · distortion 0.06 mm
→ policy.evaluate recipe inside envelope · 6 bounded writes permitted
→ executing prep.release ....... ok 48s
→ executing recoat.check ....... ok layer 412 dosing +4%
→ executing melt.stabilise ..... ok layer 1872 power −3%
! approval required hatch_change part 17 · outside envelope
→ declined AM process engineer · part 17 → CT priority
→ executing inspect.ct ......... ok 24 volumes · 11m
→ scenario complete 23 released · 1 NCR · genealogy complete
$ powdrix builds show run_7c2e4b --format genealogyLand on one printer with a measured baseline. Expand by machine, module, floor and autonomy level. Outcome components are available on build failure, yield, qualification and energy. Annual prepay carries a 15 to 20% discount.
In-situ, parameter-control or inspection AI for one printer. For manufacturers validating ROI on the wedge.
The whole loop: prep and parameter, melt and layer, recoat and powder, thermal and distortion, defect, post-process, yield and takt, robotic handling and the twin.
Multi-site AM manufacturers and bureaus standardising on Powdrix, with custom alloy, polymer and qualification models.
We have no customer quotes to show yet and will not invent any. These are the three roles on an additive floor we build for, and the problems they bring to the first meeting.
What this person is dealing with today
What this person is dealing with today
What this person is dealing with today
Powdrix reads and writes through the systems already on the floor. No rip and replace, no parallel historian, no new slicer to learn. Vendor names are examples of systems found on additive floors; connector availability is confirmed per site.
Build-preparation and slicer software (Materialise, Autodesk Netfabb, Oqton, Siemens NX AM and nTop class tools)
Geometry, orientation, supports and parameter sets in; validated recipes out
Laser and electron-beam PBF, DED and polymer SLS, MJF and FDM machines through OEM control APIs
Machine state and in-situ streams in; bounded parameter writes inside the envelope
Coaxial and off-axis melt-pool cameras, photodiodes, layer and recoater cameras, thermal sensors (EOS EOSTATE, Nikon SLM, Sigma Additive and Velo3D Assure class systems)
Frames and streams fused at the printer edge
CT scanners, surface vision and metrology (Zeiss, Nikon CT, Waygate and Volume Graphics class systems)
Volumes, images and scans in; grading, routing and packets out
Stress-relief and heat-treatment ovens, HIP units, depowdering and support removal (Solukon, AM Solutions and PostProcess class equipment)
Cycle data in; recipe parameters inside the qualified window out
Sieving, lot tracking and reuse records
Lot, reuse count and quality into genealogy
Jetson and Isaac driven handling cells, grippers, transfer rigs
Build-complete and oven signals in; handling tasks out
AM MES and build management (Oqton, Authentise, 3YOURMIND, AMFG and Materialise CO-AM class systems)
Jobs and orders in; genealogy, holds and dispositions back
Powdrix writes to production printers that make qualified parts. Every capability is scoped to the qualified envelope, every write is policy-checked, and every action is written to an append-only audit log the floor owns.
| Item | Scope | Status |
|---|---|---|
| SOC 2 Type I | Cloud control plane | QUEUED Planned, first 6 months |
| SOC 2 Type II | Cloud control plane | QUEUED Planned, 6 to 12 months |
| SSO and RBAC | Console and approvals | QUEUED Planned, 6 to 12 months |
| AS9100 and ISO 13485 assurance packages | Evidence packets and model change control | QUEUED Planned, 6 to 12 months; not a certification claim |
| Standards-cited reasoning | AS9100, ISO 13485, ASTM F42, ISO 52900 | RUNNING In the design, every answer cites its source |
| Per-tenant isolation and on-prem option | Recipes, geometry, models, memory | RUNNING In the design from day one |
One policy model, one audit trail, one benchmark across every machine on every floor, with the alloy, polymer and qualification models kept private to each site.
The questions operations directors, process engineers and certification engineers ask in the first meeting.
Only inside the OEM-approved parameter envelope, only on an explicit allow-list with per-tag limits, and only at the autonomy level your floor has set. Level 1 is advisory: Powdrix predicts and recommends and a human enters everything. Design partners start there, in shadow mode. Anything that leaves the envelope or affects qualification always waits for a named approver.
Control returns to the printer's own controller and the standing recipe. Powdrix is a supervisory layer on top of the machine controls, monitoring systems and MES you already run, never a replacement for them, so an outage degrades the floor to its current way of running, not to a stop.
Advisory recommendations start from the twin's physics and the part spec in the first weeks. Defect and build-failure prediction improve as layers, CT results and engineer corrections accumulate per part family. The first models are trained on existing in-situ camera data; synthetic rare-defect scenarios augment them only after validation against real CT.
Only if you choose cloud training. Parameter sets, geometry, defect libraries and process memory are isolated per tenant and never used to train another customer's models. Cross-customer learning is federated and IP-isolated and waits on written data agreements. A fully on-prem deployment is available for sensitive producers.
It is designed not to. Validated parameters never change without the sign-off your quality system requires, model changes go through change control, and every action carries model version, data lineage, approval status and qualification impact in the audit log. Powdrix does not claim AS9100 or ISO 13485 certification; assurance packages are planned for the 6 to 12 month release.
You are, as with any process change, which is why every write is policy-checked, bounded, logged and reversible, and why anything that affects qualification waits for a named approver. The audit log records the request, the reasoning, the limits applied and the human decision.
Start with one printer or one inspection cell and one measurable baseline. A paid pilot on melt-pool and build-failure prediction, CT porosity triage or pre-build recipe optimisation shows the number before you commit further.
↑↓ navigate↵ openesc close