NEWF/WORKS

Software & systems engineering · Huntsville, Alabama

Design, build, and verify
the system as a whole.

NewfWorks designs and builds systems that combine operational software, AI, edge computing, sensors, and prototype hardware. We stay involved from architecture through integration and test, with particular attention to the interfaces and operating constraints that determine whether the system works as a whole.

Discuss a technical problem See what we build
REFERENCE ARCHITECTURE / 01 system view
Reference architecture for a sensor-driven operational system Sensor observations flow through edge computing into operational software. Operator commands, external interfaces, and a system record show the surrounding control and evidence paths. OBSERVATIONS STATE SERVICES COMMANDS EVENTS + RESULTS INPUTS SENSORS EDGE COMPUTE SOFTWARE OPERATIONS INTERFACES EXTERNAL I/O OPERATOR CONTROL EVIDENCE SYSTEM RECORD
InterfacesDefined
BehaviorTested
EvidenceTraceable
Operational softwareAI & decision supportEmbedded computingHardware integration

01 / CAPABILITIES

Software, edge, and hardware engineered as one system.

01

Operational software

We build web applications, APIs, data services, and operator tools for environments where networks, computing resources, or deployment options may be constrained.

  • Web applications
  • APIs & data services
  • Operator tools
02

AI & decision support

We develop perception and decision-support software with defined operating limits, evaluation against specific tasks, human review where needed, and fallback behavior when automation cannot proceed safely.

  • Decision support
  • Perception
  • Automated evaluation
03

Embedded & edge systems

We build device software and sensor pipelines around real limits on timing, calibration, computing power, and connectivity. Recorded inputs can be replayed during testing and debugging.

  • Sensor processing
  • Edge compute
  • Record & replay
04

Hardware integration & test

We compare, integrate, and test sensors, processors, power systems, and packaging. The work can range from component studies and bench prototypes to hardware-in-the-loop testing and field validation.

  • Component selection
  • Hardware-in-the-loop
  • Field validation

02 / SELECTED WORK

Selected work.

We use our own products to work through problems that often make integrated systems difficult: reconstructing geometry from imperfect observations, protecting privacy across an application, operating without a network, and giving operators useful ways to review the result. Each write-up distinguishes what works today from what still needs to be validated.

Review the projects
01 / SPATIAL SYSTEMSACTIVE R&D

Mapper

Mapper turns observations from a manually driven survey into estimated landmarks, course boundaries, a centerline, and uncertainty estimates an operator can review—all without a network connection.

  • Sensor pipeline
  • Spatial reconstruction
  • Human review
02 / APPLIED PRIVACYACTIVE DEVELOPMENT

Event platform

A privacy-focused event operations platform built in Rust and PostgreSQL, with passkeys, client-side encryption, data disclosed only for a specific event, and separate encryption keys for separate purposes.

  • Rust + Postgres
  • Passkeys
  • Separate encryption keys

03 / ENGINEERING METHOD

Integration risk lives between the components.

Integration problems usually start with mismatched assumptions about timing, data, authority, resources, or failure behavior. We identify those assumptions early and test them as part of the system—not after each component appears finished.

A

Start with the operating context

Before selecting tools or components, we map the workflow, interfaces, constraints, and failure modes. That keeps architecture decisions tied to the real use case.

B

Make behavior testable

We translate requirements into behavior that can be observed and tested. The resulting evidence clearly distinguishes simulation, bench testing, and field results.

C

Design for deployment

Security, the deployment environment, data rights, component availability, and long-term support shape the design from the start—not after the prototype works.

04 / OPERATING VALUES

How our values shape the engineering.

Our values show up in concrete technical choices: what data gets collected, who retains control, how failure is handled, and what evidence supports a claim.

01

Evidence before claims

We separate measured results from simulation, inference, and assumption—and document unresolved questions alongside successful results.

02

Data minimization

We collect only the data the system needs. We also limit which records can be linked, separate access by purpose, and encrypt data as it crosses trust boundaries.

03

Operator and owner control

We make operator authority explicit and avoid unnecessary dependence on a vendor. That means documented interfaces, portable deliverables, local operation where appropriate, and documentation the owner can maintain.

04

Failure behavior by design

Before a demonstration or field test, we define who or what has authority, how the system behaves when part of it fails, and how it recovers.

05 / ABOUT NEWFWORKS

Engineers directly accountable for the work.

NewfWorks was founded in Huntsville to keep technical decisions close to implementation, integration, and test.

The founding team brings more than 15 years in software and more than 14 years in systems and electrical engineering. That experience spans commercial platforms, aerospace, defense and federal programs, nuclear power, and industrial controls—from requirements and architecture through implementation, integration, qualification, and test.

Read company background

06 / CONTACT

Have a system to build
or a technical risk to resolve?

Tell us what the system needs to do, what exists today, and where the uncertainty is.

[email protected]