All Together, Extraordinary www.adunaglobal.com Aduna is a joint venture backed by the world's largest mobile operators. We build one platform that exposes standardized network APIs (CAMARA / GSMA Open Gateway) to developers and enterprises worldwide: Our platform runs on cloud infrastructure across many regions. We ship to production every week and we are small enough that your work reaches customers fast.

The role We are building the AI & Data team. You join as one of its first engineers and you report to the Head of Engineering. This is a hands‑on role. You design, build and run AI systems in production, and you shape how AI fits into the platform.

Three things are on the table for the first year:

- Take our MCP server from proof of concept to product. AI agents will consume network APIs the same way apps do today. You build the server, the agent registration and trust model, the guardrails and the observability that go with it.
- Build the internal AI tooling our engineering and service teams use every day: code assistant and assistants over our documentation, Jira and SharePoint, automation of repetitive work, and the evaluation harness that tells us whether any of it is worth keeping.
- Build and operate the data and ML pipelines behind our analytics products and platform signals on AWS or Azure. What you’ll do - Design, ship and operate LLM and agent based services on our Kubernetes platform, with the same rigor we apply to any production API: auth, rate limits, monitoring.
- Own the production MCP server: tool definitions over CAMARA APIs, per-agent identity, risk-tiered access to sensitive APIs, threat model and anti‑fraud, anomaly detection.
- Package everything as code: Terraform, Helm, ArgoCD, CI/CD. If it is not in Git it does not exist.
- Contribute to open source and standards work around agentic access to network APIs, including our CAMARA reference implementation.
- Write short design documents, review code, mentor engineers from other teams who want to use AI in their area.

What You

Bring - 5+ years of professional software engineering experience, including ownership of production systems.

- Strong Python engineering skills.
- Hands‑on experience designing, shipping and operating at least one production LLM or agentic application.
- Deep understanding of LLM application architecture: RAG, embeddings, tool/function calling, structured outputs, context engineering, model selection and evaluation.
- Experience designing evaluation strategies for AI systems,



including regression datasets, task‑quality metrics, latency, reliability and cost.
- Understanding of AI/ML fundamentals and the ability to determine when to use LLMs, traditional ML, deterministic software or a combination of approaches.
- Experience operating containerized services on Kubernetes in AWS or Azure.
- Understanding of production AI security concerns, including authentication/authorization, prompt injection, tool misuse, secrets management, data leakage and least‑privilege access.
- Strong engineering fundamentals: testing, CI/CD, observability, code review and incident response.
- Ability to communicate architecture, technical decisions and trade‑offs clearly in English and French. Bonus - MCP or production experience with agent/tool protocols.
- Agent orchestration frameworks and multi‑agent systems.
- Azure AI Foundry/OpenAI, Amazon Bedrock/SageMaker or Databricks.
- RAG, vector databases and enterprise search.
- ML pipelines, inference, drift/anomaly detection and streaming architectures.
- OAuth 2.0/OIDC, Vault and threat modelling.
- Telecom/network APIs, CAMARA or GSMA Open Gateway.
- LLM observability and evaluation platforms.
- Open‑source contributions.

Why Join

Aduna?

- Small teams with explicit ownership. You own outcomes, not tickets.
- Production first. Every engineer is close to what runs and to the customers who use it.
- Global Teams in Montreal and Lodz (Poland), operators and customers on every continent.
- Build technology that impacts millions.
- Growth in a fast‑moving, innovative environment.
- Hybrid model (3 days in our Montréal office) - Competitive salary + full benefits (dental, medical, RRSP/DPSP) - Unlimited PTO - High ownership, collaboration, and room to grow By applying, you consent to the processing of your personal data for recruitment purposes.

Privacy notice: https://www.adunaglobal.com/privacy-notice/ Aduna is committed to a diverse and inclusive workplace and welcomes applicants from all backgrounds. Montréal, QC · Hybride Aduna est une coentreprise soutenue par les plus grands opérateurs mobiles au monde.



Nous construisons une plateforme unique qui expose des API réseau standardisées (CAMARA / GSMA Open Gateway) aux développeurs et aux entreprises partout dans le monde: Notre plateforme tourne dans le nuage, dans de nombreuses régions.

Nous livrons en production chaque semaine et nous sommes assez petits pour que votre travail atteigne les clients rapidement. Le role Nous bâtissons l'équipe IA et données. Vous en serez l'un des premiers ingénieurs et vous relèverez du Head of Engineering.

C'est un poste très technique. Vous concevez, construisez et opérez des systèmes IA en production, et vous définissez la place de l'IA dans la plateforme.

Trois chantiers pour la première année:

- Faire passer notre serveur MCP de la preuve de concept au produit. Les agents IA consommeront les API réseau comme les applications le font aujourd'hui. Vous construisez le serveur, le modèle d'enregistrement et de confiance des agents, les garde‑fous et l'observabilité qui vont avec.
- Construire l'outillage IA interne utilisé chaque jour par nos équipes d'ingénierie et de service: assistant de code et assistants sur notre documentation, Jira et SharePoint, automatisation des tâches répétitives, et le banc d'évaluation qui nous dit ce qui mérite d'être gardé.
- Construire et opérer les pipelines de données et de ML derrière nos produits analytiques et les signaux de la plateforme sur AWS ou Azure. Ce que vous ferez - Concevoir, livrer et opérer des services LLM et des agents sur notre plateforme Kubernetes, avec la même rigueur que pour n'importe quelle API en production: authentification, limites de débit, surveillance.
- Être responsable du serveur MCP en production: définitions d'outils sur les API CAMARA, identité par agent, accès par niveau de risque aux API sensibles, modèle de menace et antifraude, détection d'anomalies.
- Tout livrer sous forme de code: Terraform, Helm, ArgoCD, CI/CD. Ce qui n'est pas dans Git n'existe pas.
- Contribuer aux travaux open source et de standardisation sur l'accès des agents aux API réseau, dont notre implémentation de référence CAMARA.
- Rédiger de courts documents de conception, réviser du code, accompagner les ingénieurs des autres équipes qui veulent utiliser l'IA dans leur domaine. Ce que vous apportez - 5+ ans d'expérience professionnelle en génie logiciel, incluant la prise en charge de systèmes en production.
- Sol

📌 Senior AI Developer / Développeur(se) IA senior (Montreal)
🏢 Aduna Global
📍 Montreal