Mehdi L.

- Features development (mostly REST APIs)

- Monitoring and debugging

- Refactoring of some old code

- Maintaining the legacy code

- Spikes & PoC

- Development of the base for a stock management system (including daily and permanent stocks, stock batches, ...)

- Improvement of a blockchain of signed events to reach a French norm specifications

CallDesk is developing a Software-as-a-Service platform to manage the whole process of creation, training and release of callbots (which are voicebots adapted to phone). 🤖

Those callbots are integrated to customers services call centers and automatically handle repetitive, with low added value, phone calls, in order to help human agents focus on high value interactions and turn unsatisfied customers into brand advocates. 📲

How do they manage to achieve this ?

By using some ML and a lot of NLU, they are able to understand a human (what it wants to say, what kind of information it wants to gives, etc.) and interact with them. 🗣️

What are we doing, @CallDesk ?

- Developing the platform, making it easier to use, thus accelerating the development of callbots. 🏎️

But also conceiving new features ✨, and making sure our existant code base stay clean.

- Developing our AI platform (ASR, TTS, dialog manager, intentions understanding, etc.) and the telephony platform (that enables you and I to call a bot whenever we want). 📞

- Developing, maintaining and upgrading, customer-specific bots : creating the script, linking the bot to the customer IT infrastructure, training it, etc. 🤖

- Helping customers in the development of a bot : studying their use-case, helping them find the best approach, advising them on best practices, etc. ℹ️

What is our technical stack, @CallDesk ?

We have a strong micro-services architecture hosted in the Cloud. Most of our code base is written in JavaScript (NodeJS for the back-end, React and Redux for the front-end). We host our data on severals NoSQL databases ans some SQL ones. We have a solid CI process, which is near reaching the CD step.

We managed to reach a 100% unit-tested code base, which means each lines of code is tested (and deployed to some test environments) before being deployed to a production environment. ✅