Job Description

A little about us

SPOTIO is a dynamic, fast-growing American start-up with a 10-year tradition of creating the #1 Sales Engagement Platform. SPOTIO's platform helps field sales teams manage sales activities, increase the productivity of sales representatives and record field sales insights.

We have offices in Dallas, Texas and Gdansk. In Gdansk, there is a development, product and QA team totalling 21 people (50 people in total globally).

How do we work?

There's no corporate vibe or dress code here. Instead, there's a friendly, collaborative atmosphere and small teams. On Tuesdays and Wednesdays, we meet in the office in Gdansk because we want to. On other days, we work remotely because we can.

The team works normal local business hours and we have calls with the Dallas team several times a week, typically only during overlapping working hours of 2:00pm to 5:00pm Polish time. There is also an opportunity for travel to Dallas, however this is not required.

About the Role

As SPOTIO's AI/ML Engineer, you will take full ownership of the direction and execution of our machine learning and AI initiatives. You will maintain our existing production models and infrastructure while architecting new ML systems that power intelligent recommendations, predictive scoring, and AI-driven workflows across the entire SPOTIO platform.

You will own the end-to-end ML lifecycle: feature engineering, model training and evaluation, production deployment, drift monitoring, and retraining operations. You will also work closely with our engineering and product teams to integrate ML outputs into our LLM layer, shape the AI descriptions and recommendations that reps see in the product, and potentially fine-tune large language models as that capability matures.

Current State and Handover Context

SPOTIO's ML platform is already live in production. The initial build (containerized scoring service, gradient boosting models, feature engineering pipelines, batch retraining, drift monitoring) was delivered by an external consulting team that has since rolled off the engagement. You will inherit a well documented, operational system rather than building from scratch. There will be opportunities to work with the consultants during the onboarding process.

Comprehensive onboarding material is in place, covering system architecture, model training procedures, feature pipelines, deployment workflows, and operational runbooks. Your first focus will be absorbing that documentation and taking operational ownership of the existing platform.

Role Focus and Priorities

This is fundamentally an ML Engineering role with strong MLOps responsibilities. You will spend most of your time owning, evolving and operating production ML systems, not building research prototypes.

Priorities, in order:

1. Operate and maintain the existing production ML platform (scoring service, retraining cycles, monitoring).
2. Evolve the platform: improve model quality, expand feature engineering, create custom models, add new ML capabilities as the data and product mature.
3. Integrate ML outputs into SPOTIO's LLM-powered AI layer alongside the engineering team.
4. Explore advanced approaches (reinforcement learning, LLM fine-tuning) as the platform matures.

What Success Looks Like (First 3-6 Months)

By the end of your first six months, you should have:

• A complete working grasp of SPOTIO's ML and data pipeline: scoring service, feature pipelines, training workflows, model registry, and deployment process.
• Confidence monitoring all production models and the ability to diagnose and resolve scoring failures or batch issues independently.
• The ability to retrain and promote new model versions through the established quality gates without external help.
• Strong working knowledge of SPOTIO's underlying data model (customers, activities, data objects, custom fields) and how it flows into ML features.
• An informed point of view on what to improve next, grounded in what you've actually seen in production.

Team and Reporting

You will report directly to the CTO (based in Dallas). Day-to-day you will collaborate with:

• The Gdansk engineering team (development squads and QA).
• Dallas-based product and engineering leadership.

Requirements

Your Main Responsibilities

• Own the full ML lifecycle for SPOTIO's AI features: problem framing, feature engineering, model training and evaluation, production deployment, and ongoing retraining as new data accumulates.
• Operate and evolve the production scoring service, a containerized REST API that serves real-time predictions across multiple ML capabilities for all active SPOTIO customers.
• Monitor model health across all companies and capabilities: track retrospective accuracy metrics, feature drift statistics, nightly batch success rates, and API latency through SPOTIO's observability stack.
• Manage the model retraining cycle, including assembling training datasets, evaluating candidate models against quality gates, and promoting new versions through the model registry.
• Triage and resolve production scoring failures, including upstream data access issues, feature extraction errors, and batch pipeline anomalies.
• Manage ML feedback loops, including reinforcement-style approaches such as Thompson Sampling bandits, to incorporate user behavior signals into model training over time.
• Collaborate with SPOTIO engineering to integrate ML outputs into the LLM-powered AI layer, shaping the AI-generated descriptions and next-best-action recommendations that sales reps see in the product.
• Work with product and engineering on the design and activation of new AI features beyond the current scoring suite.
• Contribute to SDLC ceremonies: architecture design, code review, CI/CD, and agile planning.

Requirements

• At least 5 years of professional experience in ML engineering, applied data science, or a closely related role.
• Strong Python skills, with experience building and deploying production ML services (FastAPI or equivalent).
• Hands-on experience training, evaluating, and deploying gradient boosting models (LightGBM, XGBoost, or similar) in production.
• Experience with model explainability tools, particularly SHAP.
• Infrastructure-as-code experience, ideally with Terraform (all SPOTIO infrastructure is deployed via Terraform).
• Working knowledge of MLOps practices: model registries, versioned artifacts, drift monitoring, and automated retraining pipelines.
• Proficiency in SQL with an understanding of query patterns, connection pooling, and performance considerations in high-volume environments.
• Experience with PyTorch, scikit-learn, NumPy, and Pandas (or Polars).
• Precision, proactivity, and comfort taking full ownership of systems.
• Proficiency in English.
• Education related to computer science, mathematics, statistics, or a related field.

Nice to Have

• Experience with multi-armed bandits or reinforcement learning approaches.
• Familiarity with Elasticsearch.
• Azure Event Hubs, Stream Analytics, or ADLS Gen2.
• Production experience with a major cloud platform (Azure, AWS, or GCP).
• Experience with Kubernetes in a production environment.
• Azure DevOps Pipelines or Argo CD.
• Exposure to sales CRM data, territory management, or field sales workflows.

On-Call Expectations

Because the scoring service runs in production and serves real-time predictions to all SPOTIO customers, this role carries on-call responsibility for ML-related incidents. The platform is stable and well-instrumented, so out-of-hours incidents are expected to be rare. On-call rotation is shared and supported by the broader engineering team.

Tech Stack You'll Work With

ML and Data Science

Python, LightGBM, SHAP, Thompson Sampling, feature engineering, model registries, LLM integration and fine-tuning

Cloud and Infrastructure

Microsoft Azure (SQL, Cosmos DB, Blob Storage, ADLS Gen2, Event Hubs, AKS, Container Apps, Stream Analytics)

API and Services

FastAPI, REST APIs, Managed Identity / API key auth, VNet-internal services

Data and Storage

Azure SQL (multi-zone), Cosmos DB (MongoDB API), Elasticsearch, Redis, Azure Blob / ADLS Gen2

DevOps and Monitoring

Azure DevOps Pipelines, Argo CD (planned), Application Insights, Azure Monitor, Docker, Kubernetes

Benefits

What we can offer you:

• Interesting work and real impact on the product, organization and technology selection
• Modern equipment
• Working with great people ;)
• Free parking
• Modern office
• Medicover sport and health
• Paid vacation, sick leave
• Flexible working hours

Location: Poland (R&D Team), Gdańsk

Type of work: hybrid (2 days a week from a CUBE office)

Contact person: Agnieszka Myśliwczyk