Softwise builds ERP integrations, AI workflow automation, mobile apps, and SaaS platforms for clients in France. The work is done by 20+ in-house developers covering iOS, Android, data, and AI, with no subcontracting, delivered remotely from the head office in Tunis in French, English, or Arabic. France is one of the markets behind Softwise's 100+ delivered clients, alongside Tunisia, Saudi Arabia, United Arab Emirates, and Canada.
The same catalogue as every other market, delivered by the same in-house team of 20+ developers.
ERP integrations, custom CRM builds, and Salesforce or HubSpot wired into the systems around them. An integration layer means each system integrates once, so replacing one of them later does not break the others. Data engineers sit on the same team as the application developers.
Automation and AI added between an inbox, a form, or a queue and the system of record you already use, writing through the interfaces your staff use today. The judgment stays with people: extraction is reviewed or sampled before it commits, and accuracy is measured before anything is trusted.
SaaS platforms covering authentication, roles, billing, dashboards, and multi-tenancy, plus native iOS and Android applications built by in-house mobile engineers. Work ships in tested increments, so the product can be steered while it is being built.
The parts of the regulation that change how a system is built, and how we approach them.
The GDPR governs how personal data is collected, stored, and transferred, including outside the EU. That is answered with architecture rather than a policy document: which region the database runs in, what crosses a border, what a backup replicates, and which third-party service sees what. The decisions are made with you and written down, so where your data lives is a fact rather than a guess.
Access, rectification, erasure, and portability are requests your system has to be able to answer. Built in from the start they are ordinary features: find everything about one person, export it, correct it, delete it. Retrofitted into a schema that never planned for them, the same requests turn into manual database work every time one arrives.
Data minimisation is a design constraint, not a compliance checkbox: fields that are not collected cannot leak, and logs that do not record personal data do not need a retention policy of their own. Retention windows and what gets written to application logs are decided during the build, with you, not discovered during an audit.
Two French engagements: an ERP integration that ended double data entry, and AI put inside a claims workflow without replacing it.
Softwise connected a Lyon distributor's ERPNext instance to its e-commerce storefront and third-party logistics provider. Orders, stock, and invoices now flow automatically in both directions. Staff re-keyed every order before the integration; after it, orders move with no manual entry and stock accuracy on the storefront went from day-old to near real time.
Read the case study →/005/Softwise integrated AI document extraction into a Paris insurance firm's existing claims process without replacing any system the staff used. Incoming claim documents are classified and extracted by a model, verified by staff in a review screen, and pushed into the same back office as before. Every AI output is checked or spot-checked by a human before it touches a claim, and intake that took days now completes in hours.
Read the case study →A B2B distributor near Lyon selling through a web storefront, with ERPNext running purchasing, inventory, and invoicing, and a third-party warehouse shipping the orders.
Re-keyed orders eliminated, stock on the storefront near real time instead of day-old
ERPNext (Frappe framework), e-commerce storefront, 3PL warehouse API
An insurance services firm in Paris processing claims at volume. Every claim arrived as a bundle of PDFs and photos by email. Staff read each document, typed fields into the claims system, and routed the file to the right handler.
Intake cut from days to hours, staff hours moved from transcription to claim handling
AI extraction with human review, shadow-mode validation before go-live
Working with us
Language, hours, and travel, stated plainly rather than implied.
Requirements, reviews, and day-to-day questions run in French, with the same in-house engineers who are building the product. English and Arabic are available on the same team, so a mixed-language stakeholder group is not a problem.
Tunis matches Paris in winter and runs an hour behind it in summer, so a French working day and a Tunis working day are effectively the same day. Calls, reviews, and answers land inside your hours rather than a message behind them.
Remote by default, with travel to your location when a project calls for it. One office, in Tunis, and no office in France.
Yes. France is one of the markets Softwise has delivered in, out of 100+ clients across Tunisia, France, Saudi Arabia, the UAE, and Canada. Delivery is remote from the head office in Tunis, in French, English, or Arabic, with travel to France when a project calls for it.
Yes. Softwise delivers in French, English, and Arabic. Requirements, specifications, reviews, and day-to-day communication run in French with the same in-house engineers who are building the product. French-language interfaces are scoped as part of the build rather than treated as a translation task added at the end.
Data location is an architecture decision made with you, not a default: which region the database runs in, what crosses a border, what backups replicate, and which third-party services see what. The GDPR governs how personal data is collected, stored, and transferred, so those choices are documented against your requirements and reviewed with you before launch.
Yes. Softwise connected a Lyon distributor's ERPNext instance to its e-commerce storefront and third-party warehouse, so orders, stock, and invoices flow automatically in both directions. A middleware layer owns every flow, which means each system integrates once and replacing one of them later leaves the others untouched.
Yes. The pattern is an AI layer between an input and the existing system of record, writing through the interfaces the current process already uses. Softwise built exactly that for a Paris insurance firm: documents are extracted automatically, a person approves each one, and the claims system never changed.
No. Softwise has one office, in Tunis, Tunisia, and no French entity, SIRET, or local phone number. Delivery is remote, in French or English, with travel to France when a project calls for it. Tunis matches Paris time in winter and is an hour behind in summer, so the working day overlaps almost entirely.
The same in-house team delivers all five.
Native iOS and Android from developers who ship them weekly. Store release and post-launch fixes included.
Read more →/002/Auth, billing, dashboards, and an architecture that survives your next 10x in users. Shipped in tested increments.
Read more →/003/Data engineers on the same team as the app developers. The integration is designed in, not bolted on.
Read more →/004/Custom CRM builds, and Salesforce or HubSpot wired into the systems your team already uses. Your pipeline and your product data stop disagreeing.
Read more →/005/We automate what your people repeat, then add AI where it moves a number you already track.
Read more →