Now booking software engineering engagements for Q3 — Q4 2026

Unified commerce for operators
who already have the data.

Unified-commerce platforms across storefront and channel, customer data platforms that don't pretend to be something they're not, merchandising and pricing AI, and the data consolidation that makes the inventory number agree across every system that asks.

The landscape

What retail software looks like in practice.

Retail operators run a stack that has accreted over a decade: a commerce platform that picks up new channels every year, a POS that came from a different vendor, an OMS that joined them later, a WMS that the operations team picked, a CDP that was supposed to solve customer identity, and a data warehouse that everyone treats as the source of truth except when they don't.

The pressure isn't usually to replace the commerce platform. Replatforming is expensive and the migration risk is real. The pressure is on the surrounding work: the channel integrations that drop orders on bad days, the inventory number that disagrees between the POS and the OMS, the customer profile that exists in three CDPs at once, the merchandising decisions that still happen in spreadsheets. AI is starting to land here in real ways, but the underlying data has to actually be clean enough for the AI to mean something.

Apollo builds the unification, data, and AI work that sits around the commerce stack. Channel-aware by default, identity-aware for the customer layer, and operator-aware for the parts buyers and store managers actually open.

The regulatory reality

PCI DSS for payment handling, the state privacy stack (CCPA, CPRA, and the growing list behind them), GDPR for any operator with EU customers, ADA for the storefront and digital channels, and the consumer-protection rules each state attorney general enforces. Specialty categories add their own: TCPA for marketing, COPPA for kid-facing sites, FDA for some consumables.

Where systems break

The inventory feed that's two minutes behind the floor on a busy Saturday. The OMS that thinks the order shipped but the WMS knows it didn't. The customer record that exists in the CDP, the ESP, the loyalty system, and the POS, all slightly different. The reporting that takes 18 minutes to refresh because the data warehouse joins are pathological. The merchandising spreadsheet a senior buyer has maintained since 2018.

Where AI lands

Demand-forecasting and replenishment models that use the data that's actually clean. Personalization that doesn't violate privacy boundaries. Pricing and markdown decisions with explicit reasoning and the operator override paths built in. Merchandising copilots that pull sales, returns, and cohort performance at the same time. Customer-service AI that knows the order history and respects the channel.

Where we work

Four engagement patterns in retail.

Most engagements with retail operators land in one of these shapes. Each gets built around the commerce stack you already run, with the data and integration discipline a multi-channel operation actually requires.

Unified Commerce Platforms

The integration layer between commerce platform, POS, OMS, WMS, and the channel-specific systems. Order routing that respects fulfillment economics. Inventory unification that finally lets the storefront and the warehouse agree. Built around the platforms you already run, not as a forklift replacement.

Channel integrationOrder routingInventory unificationPlatform-adjacent

Customer Data Platforms

The customer-identity, profile, and consent layer that supports personalization, loyalty, and the rest of the customer-data stack. Identity resolution across channels, consent tracking that respects state privacy law, and integration with the systems marketing and service actually use. Built when an existing CDP isn't doing the job a CDP is supposed to do.

Identity resolutionConsent managementProfile unificationChannel integration

Merchandising & Pricing AI

Demand-forecasting, replenishment, markdown, and price-optimization models built on the operational data that's actually clean. Buyer and pricing copilots that pull sales, returns, and cohort performance together. Every recommendation attributable, every override logged, every model versioned.

Forecast / replenishMarkdown / pricingBuyer copilotOverride paths

Operations & Inventory Data

POS, OMS, WMS, returns, and supplier data unified into a single query-able layer. Identity resolution on the customer side, SKU and lot resolution on the operations side, and the lineage required for reporting that finally agrees with itself. The reporting layer and the analytics layer see the same numbers.

POS + OMS + WMSSKU resolutionReturns and supplierReporting parity
The systems landscape

Where our work sits.

Retail operators run a small set of large platforms across commerce, POS, OMS, WMS, and customer data, with a long tail of channel and supplier integrations. Apollo builds the unification, data, and AI work that sits around this landscape, not on top of it. The map below is simplified, but representative of what we walk into.

COMMERCE & OPERATIONSCHANNEL & CUSTOMER SOURCESAPOLLO LAYERAPPLICATIONS + AICommerce / StorefrontWeb · mobile · headlessPOS / Store SystemsStores · returns · associate toolsOMS / WMS / InventoryOrder routing · warehouse · stockChannel FeedsMarketplace · social · partnerCustomer SignalsWeb · email · service · loyaltySupplier & VendorPOs · ASNs · catalog · returnsIntegration LayerAPIs · events · file feedsApollo Platform LayerSKU · identity · lineageCustomer Data LayerProfile · consent · resolutionUnified Commerce UIsOrder · inventory · returnsMerch / Pricing AIForecast · replenish · markdownBuyer & Planner ToolsCohort · margin · returns viewService / PersonalizationChannel-aware · consent-aware
Cross-cutting concerns
SKU and lot resolutionCustomer identity & consentPCI DSS scope boundaryChannel-by-channel privacy logicModel versioning & overrides
Sits around the commerce stack, not on topChannel-aware by constructionPrivacy-aware by defaultOperator UIs are first-class
The regulatory context

Compliance, designed in.

The frameworks that shape what we build, and what we ship.

Retail lives under a privacy, payment, and consumer-protection stack that's gotten meaningfully more complicated in the last five years. The right set of frameworks for any engagement depends on the geographies in scope, the channels in use, the categories sold, and the customer audiences served.

Apollo treats these as design inputs, not as a checklist applied at the end. Payment-data scope boundaries, consent management, identity resolution, and accessibility get specified before the first line of code and reviewed against the relevant frameworks at every iteration.

The panel on the right is the working set we encounter most often. Any given project picks a subset, and the proposal will be explicit about which ones apply and what we will and won't certify directly.

retail / compliance-frameworks

Payment & commerce

PCI DSSPayment-card data handling, segmentation, and scope reduction
ADA / WCAGAccessibility for digital storefronts and store-facing systems
FTCFederal Trade Commission rules on advertising and consumer protection
State AGState attorney-general consumer-protection enforcement

Privacy & data

CCPA / CPRACalifornia consumer privacy and rights-management rules
State privacyVCDPA, CPA, CTDPA, and the growing patchwork of state laws
GDPRWhere European customers or operations are in scope
TCPA / CAN-SPAMMarketing-channel consent and opt-out rules

Category & controls

COPPAChildren's online privacy where kid-facing sites are in scope
FDA / FTC categoriesCosmetics, supplements, and regulated-category labeling
SOC 2 Type IIControls attestation for the platforms we build
How we engage on retail work

Four phases. Built around the operator.

Apollo's standard methodology, applied to retail. Channel topology, identity strategy, and the data discipline a multi-channel operation requires get specified before the build starts, so the storefront and the back office can finally agree.

Discovery

Map the channels. Map the data.

The current platforms (commerce, POS, OMS, WMS, CDP), the channel integrations, the volumes, and the breakpoints your team already knows about. The categories sold, the geographies in scope, and the regulatory frameworks they bring. Where AI helps, where rules suffice, and where a buyer still has to confirm.

Design

Architecture, identity plan, data plan.

Integration topology across the commerce stack. Identity resolution and consent design for the customer layer. SKU and lot resolution for the operations layer. Operator-UI wireframes for buyers, planners, and store managers.

Build

Platform. Data layer. Operator UIs.

Integration layer deployed, channel and partner feeds connected, customer data platform wired up with consent tracking, operator UIs functional. Two-week iterations. Each shipped capability arrives with its tests, its monitoring, and the lineage fields the data team will need.

Operate

Measure. Tune. Expand.

Inventory parity, order-routing accuracy, identity-resolution rates, AI suggestion-accept rates, and channel-feed uptime in production dashboards. Threshold tuning based on what the data shows. Gradual rollout to adjacent categories, channels, or banners once the first set is steady. Knowledge transfer to your team along the way.

Start a conversation

Tell us what you're trying to build.

Send a paragraph about the project: the platforms involved, the channels and categories, the volumes if you have them, and the parts that aren't working today. We'll reply within one business day, either with a 30-minute call or with an honest "this is not the right fit; here's who you should call instead."

A senior engineer reads every inquiry before anyone replies.
Thank you. Your message is in. We'll be in touch within one business day.