Book a demo
FIOS · Food Innovation Operating System

The AI platform for food product development.

Predict cost, sensory, nutrition, and regulatory risk — simultaneously, before any pilot. Know what your product will do before you make it.

faster product launches
32%lower formulation cost
< 10 minfeasibility & risk assessment
The AI platform for food product development.

Most formulation failures are visible in the data, just too late. FIOS moves that visibility to day one, so product decisions are based on what will happen, not what already has.

FIOS workflow graphThe FIOS workflow

Connect the dots of food innovation.

FIOS gives R&D, regulatory, and commercial teams one shared way to move products forward, with the same data, the same models, and no silos.

Discover

Discover

Consumer & market insight
Scientific summaries
Design

Design

Concept development
Feasibility & business case
Ingredient discovery
Develop

Develop

Formulation optimization
R&D product development
Regulatory compliance
Produce

Produce

Manufacturing scale-up
Supply chain setup
Production cost modeling
Commercialize

Commercialize

Go-to-market strategy
Launch execution
Post-launch tracking
Capabilities

Built for the product questions CPG teams actually ask.

FIOS brings seven core capabilities into a single connected workflow. Each is grounded in what global R&D teams told us they needed most.

01 — MULTICOMPONENT FORMULATION

Design products as systems, not single recipes.

Binder + particulates + coating. Wafer + chocolate + filling. One model.

Most real products are multi-component: a chocolate bar has chocolate, wafer, and filling — each with its own formulation, structure, and water activity. FIOS models each component and the interactions between them, optimizing the whole product simultaneously for sensory, nutrition, cost, and stability. Water migration between phases. Structural interactions. Scale-up behavior. Modeled together, not in isolation.

"When we make a chocolate bar, we make the chocolate on its own, the wafer, and the filling cream. We need to modulate each of these separately to give the whole nutritional package for the bar."

— Principal Technologist, global confectionery
Design products as systems, not single recipes.
02 — PROCESS OPTIMIZATION

Close the gap between bench and plant.

Formulation decisions inform processing. Processing decisions inform formulation.

Most formulation failures happen at the handoff between R&D and manufacturing — when a recipe that worked in the lab breaks at scale. FIOS links formulation and process parameters in a single model, so extrusion settings, oven profiles, and fermentation conditions are optimized alongside ingredient choices. The result: fewer pilot failures, faster scale-up, and production runs that behave the way the bench predicted.

Deployments reduce breakage and optimize shape consistency in extruded snacks — linking extrusion parameters to texture outcomes in a single predictive workflow.

— Snack, and bakery manufacturers
Close the gap between bench and plant.
03 — NUTRIENT, SENSORY & COST

Optimize for the three things every product has to deliver.

Hit your Nutri-Score target while maintaining taste and protecting margin.

Every CPG decision trades off nutrition, sensory, and cost. Reformulate for a lower sugar target and texture shifts. Swap to a cheaper ingredient and flavor profile moves. FIOS models all three dimensions together — predicting sensory outcomes from composition changes, checking nutrient profiles against Nutri-Score, HSR, and FDA Healthy criteria, and simulating ingredient substitutions with cost and performance deltas side by side.

Used to balance nutritional targets with processability, cost, and sensory quality across international Nutrient Profiling Models for global confectionery and snack portfolios.

— Fortune 500 CPG
Optimize for the three things every product has to deliver.
04 — REGULATORY COMPLIANCE

Clear regulatory risk before you commit to a product.

Model key frameworks including Nutri-Score, HSR, FDA and EFSA health claims.

Regulatory surprises are a costly failure point in CPG. When a product passes R&D but fails a regional nutrient profiling model, it can trigger months of rework. FIOS identifies regulatory risks early and validates formulations against market-specific rules. FIOS supports EFSA health claim submissions, built to accelerate and streamline dossier preparation. FDA GRAS and food additive petition formats are planned additions.

Clear regulatory risk before you commit to a product.
05 — DATA INGESTION & INTEGRATION

Your data and PIPA’s data, working together.

Proprietary recipes, supplier TDS, and production runs, with public data added when it helps.

FIOS works with the data formats you already use, including recipes in Word, PDF, or Excel, supplier technical data sheets, raw material shelf-life records, and manufacturing process logs. It can also integrate with the PLM systems your team prefers. Your data remains yours. Public data and PIPA’s curated knowledge bases can expand what your team is able to analyze and reason over. They are used together when it adds value, and kept separate when that is the right choice.

Deployed inside enterprise environments at global CPG and ingredient companies — ingesting proprietary recipe databases, and regulatory constraints alongside PIPA's scientific knowledge graph.

Your data and PIPA’s data, working together.
06 — ENTERPRISE CONTROLS

Built to fit how global R&D teams actually work.

Permissions, project isolation, audit trails, live dashboards.

FIOS is built for enterprise R&D environments, with role-based access control, project-level permissions, cross-functional workspaces, and live dashboards for different types of users. Every formulation, experiment, and result is versioned. Teams can compare iterations, track changes, and roll back when needed. From bench through manufacturing, FIOS gives everyone who needs it access to the same source of truth.

Requested by global enterprise R&D teams where R&D, regulatory, and manufacturing span continents and need shared visibility without duplicating work.

Built to fit how global R&D teams actually work.
What powers FIOS

Hybrid models: AI, physics, and chemistry working together.

Most AI-for-food platforms rely mainly on machine learning. Most simulation platforms rely mainly on physics. FIOS combines four modeling approaches into a single prediction, helping it solve problems where either approach on its own can fall short.

01

AI, ML & knowledge graphs

Pattern recognition across millions of recipes and scientific papers. Learns what works from data humans can't hold in their heads at once.

02

CFD & multiphysics

First-principles simulation for processes such as extrusion, baking, and fermentation, covering heat transfer, mass transfer, phase change, and chemical reactions as connected parts of the same system.

03

Chemical, kinetic & omics

Reaction kinetics, degradation pathways, and omics-level ingredient data show what a formulation is likely to do in practice, not only what goes into it.

04

Production data

Continuously calibrated against real manufacturing runs. Each production batch improves the next prediction.

Go deeper on the technology
Enterprise deployment

Built for how global R&D actually operates.

Security, deployment, and integration are built in from the start. FIOS already operates inside enterprise environments at some of the world’s most demanding CPG companies, meeting the procurement standards that come with them.

SaaS or on-premise deployment
Hosted in PIPA’s secure cloud, or deployed inside your infrastructure for the strictest data-residency requirements.
Role-based access & SSO
Permission controls at project, workspace, and dataset level. Integrates with enterprise identity providers.
Data ingestions & integration
Connects to own techstack and internal databases. Your systems of record stay your systems of record.
Versioning & audit trails
Every formulation, experiment, and result is stored centrally with full version history and change tracking.
Know what your product will do before you make it.
Go from trial-and-error, late surprises, and long cycles to predictable outcomes, fewer pilots, and confident scale-up.