Data Scientist | AI & ML Ops

About Addi

We are a leading financial platform, building the future of payments, shopping, and banking—a world where consumers and merchants can transact effortlessly and grow together. Today, we serve over 2 million customers and partner with more than 20,000 merchants, making Addi Colombia’s fastest-growing marketplace.

With a state-of-the-art, technology-first approach, we provide banking solutions (deposits, payments, unsecured credit) and commerce services (e-commerce, marketing), bridging the financial gap for millions and redefining how people experience financial freedom. As the country’s leading Buy Now, Pay Later provider, we have secured regulatory approval to operate as a bank, unlocking even greater opportunities for our customers. In the past year, we have also achieved profitability, reinforcing the strength of our business model and our ability to scale sustainably.

Our mission has earned the trust of world-class investors, including Andreessen Horowitz, Architect Capital, GIC, Goldman Sachs, Greycroft, Monashees, Notable Capital, Quona Capital, Union Square Ventures, Victory Park Capital, and more, who back our vision for the future. With their support, we are not just growing—we are transforming Latin America’s financial ecosystem and shaping the next generation to shop, pay, and bank in Colombia.

But what truly sets us apart is how we build. We are a conscious company, driven by deep experience in scaling technology, services and products, and we live by our values every day.

About the Role

This is where you come in. Below, you’ll find what this role is all about—the impact you’ll drive, the challenges you’ll tackle, and what it takes to thrive at Addi. If you’re ready to be part of something big, keep reading.

What’s the mission you’ll drive

Build and scale AI-driven solutions that measurably increase efficiency across Addi. You will automate high-impact workflows by designing agentic systems and hybrid ML models that run reliably in production, partnering with Engineering to turn complex business needs into scalable, automated "augmented teams.

What you will do

• Design and Build AI/ML Solutions: Design, implement, and iterate on AI and ML solutions that automate or augment high-impact workflows, selecting the right approach for the problem (classical ML, LLM-based methods, or hybrids).

• Production Delivery and Deployment: Partner with engineering to integrate solutions into production systems, ensuring they are scalable, reliable, and easy to operate and evolve over time.

• Build and Maintain Connectors and APIs: Create and maintain connectors/APIs that integrate AI/ML solutions with internal systems, tools, and data sources, including working with asynchronous execution patterns when needed, in collaboration with Engineering/ML Engineering on production constraints.

• Agent Logic and Tooling Ownership (DS scope): Own the agent logic and behavior in production (prompting, routing, tool use, guardrails, evaluation loops, and iteration), partnering with ML Engineering for the underlying serving infrastructure (e.g., Kubernetes, heavy CI/CD, platform reliability).

• Evaluation, Monitoring, and Continuous Improvement: Define success metrics, build evaluation methods (offline + online), monitor performance in production, and iterate to improve quality and reduce regressions.

• Knowledge and Data Assets: Build and maintain the required knowledge/data assets (e.g., KBs, datasets, taxonomies, retrieval/evidence structures) and keep them aligned with business and product changes to reduce failures.

• Cross-functional Collaboration: Work with stakeholders to understand the business problem deeply, clarify requirements, align on tradeoffs, and translate needs into clear scope, milestones, and measurable outcomes.

• Experimentation and Impact Measurement: Design and run experiments when needed (A/B tests, holdouts, phased rollouts) to quantify impact, validate improvements, and guide prioritization.

What we’re looking for

• Proven Technical Tenure in Applied AI/ML

  • 3+ years of experience building and deploying AI/ML solutions end-to-end, specifically in high-impact or internal automation roles.

  • Evidence of shipping models with guidance in a collaborative environment, moving beyond local notebooks into production systems.

  • Bachelor’s degree in Physics, Mathematics, Statistics, Economics, or Computer Science, providing the first-principles thinking needed for complex AI debugging.

• Possesses Experience with LLM-Based Solutions

  • Demonstrated success in building or contributing to systems that utilize modern LLM approaches (e.g., LangChain, LangGraph) to solve real-world tasks.

  • Experience designing knowledge bases or retrieval structures to improve the reliability of AI outputs.

• Possesses Strong Classic Data Science Foundations

  • Demonstrates a deep understanding of statistics, experimentation (A/B testing, sampling), and classical ML methods to ensure AI isn't used where a simpler model suffices.

  • Proven ability to frame business problems into data science solutions, moving from raw data to production-ready iteration.

• Has Solid Expertise in Deep Learning & Transformers

  • Skilled in neural architectures and optimization, with a working knowledge of attention mechanisms and transformer-based models.

  • Proficiency in modern frameworks like PyTorch, TensorFlow, or Scikit-learn to build and evaluate models from the ground up.

• Experienced in Modern AI & Agentic Systems

  • Hands-on experience building with LLMs using advanced techniques: prompting, structured outputs, tool use, and guardrails.

  • Familiarity with orchestration patterns (routing, memory, handoffs) and retrieval-augmented generation (RAG) to build reliable, non-hallucinatory agents.

• Demonstrates Ability to Build Production-Adjacent Code

  • Mastery of Python for creating reproducible pipelines and evaluation tooling.

  • Comfortable working in shared codebases using Git/GitHub, with the ability to build internal automation utilities and connectors/APIs.

• Track Record of Driving Efficiency & Impact

  • Focused on "Results over Research"—prioritizes automation rate, time saved, and throughput improvements over purely academic model performance.

  • Takes accountability for the success of projects, ensuring solutions meet both technical stability and business expectations.

• Communicates Technical Tradeoffs with Clarity

  • Exceptional ability to explain the limitations and risks of AI to non-technical stakeholders, ensuring expectations are managed and scope is realistic.

  • Proactively collaborates with Engineering to navigate production constraints, such as latency, cost, and reliability.

Why join us?

• Work on a problem that truly matters – We are redefining how people shop, pay, and bank in Colombia, breaking down financial barriers and empowering millions. Your work will directly impact customers' lives by creating more accessible, seamless, and fair financial services.

• Be part of something big from the ground up – This is your chance to help shape a company, influencing everything from our technology and strategy to our culture and values. You won’t just be an employee—you’ll be an owner

• Unparalleled growth opportunity – The market we’re tackling is massive, and we’re growing faster than almost any fintech lender at our stage. If you’re looking for a high-impact role in a company that’s scaling fast, this is it.

• Join a world-class team – Work alongside top-tier talent from around the world, in an environment where excellence, ownership, and collaboration are at the core of everything we do. We care deeply about what we build and how we build it—and we want you to be a part of it.

• Competitive compensation & meaningful ownership – We believe in rewarding our talent. You’ll receive a generous salary, equity in the company, and benefits that go beyond the basics to support your growth.

How the hiring process looks like

We believe in a fast, transparent, and engaging hiring experience that allows both you and us to determine if there's a great fit. Here’s what our process looks like:

• Step 1: People Interview (30 min)
  A conversation with a recruiter or hiring manager to get to know you, your experience, and what you're looking for. We’ll also share more about Addi, our culture, and the role.

• Step 2: Initial Interview (45-60 min)
  A more in-depth conversation with the hiring manager, where we explore your skills, experience, and problem-solving approach. We want to understand how you think and work.

• Step 3: Take Home Challenge (5-6 days)
  Complete a simple take-home challenge within a 1-week window. With this technical challenge, we want to see your technical expertise solving a real-world problem. We expect that you invest 5 hours or less in developing a working solution.

• Step 4: Take Home Challenge Review (60 min)
  Meet with a Data Scientists and the Data Science Lead to talk about your take-home exercise submission and any questions you might have.

• Step 5: Co-Founder Interview
  If there’s a strong match, you’ll have a final conversation with our Founder to align on expectations, cultural fit and ensure mutual excitement. From there, we’ll move quickly to an offer and discuss next steps.

We value efficiency and respect for your time, so we aim to complete the process as quickly as possible. Our goal is to make this experience insightful and exciting for you, just as much as it is for us. Regardless of the outcome, we are committed to always providing feedback, ensuring that you walk away with valuable insights from your experience with us.