About Popular Power

Popular Power is the intelligence layer for solar and storage. Founded in 2021 by a team with deep experience building and operating solar businesses, we build software that helps solar companies operate and scale in a rapidly changing energy system. Today, more than 500 MWp of distributed solar assets are managed on Popular Power across multiple markets.

As the energy transition accelerates, distributed assets are becoming critical infrastructure. Popular Power gives operators the clarity to grow with confidence as portfolios expand, systems become more complex, and the stakes rise. Tomorrow must be distributed today..

We’re a fast-growing, focused team that values talent density, curiosity, and high standards. If you’re excited about building category-defining software in a sector that urgently needs it, we’d love to meet you.

🔎 Who we're looking for

We're seeking an exceptional Data Scientist who excels at rapid prototyping to be a key driver of our next phase of growth. With strong product‑market fit and exciting initial traction, we're ready to scale our impact across the solar industry. You will turn ambiguous, data‑rich problems into MVP ML models that validate product hypotheses fast and unlock new feature projects on the Popular Power platform. You’ll collaborate closely with Account Managers, Design, and Engineering to frame the smallest viable experiment, test and iterate quickly based on real‑world feedback and turn winner proposals into production worthy solutions.

Popular Power is a startup! We are hiring for person, not profile. As a key early hire, you'll help shape our data foundations and modeling practices while embodying our core values of being data‑driven and solving for outcomes. If you're passionate about clean energy and love converting messy datasets into lightweight, testable models that move product forward, we want to hear from you.

👩🏽‍💻 What you'll do

Rapidly prototype ML models (classification/regression, time‑series forecasting, anomaly/event detection, clustering) that prove or disprove product hypotheses.
Drive problem framing: define success criteria, baselines, and iteration plans for MVPs with cross‑functional partners.
Perform exploratory analysis and feature engineering across time‑series, tabular, geospatial, and IoT/SCADA datasets.
Stand up lean data pipelines for ingestion and prep (SQL + Python); automate just enough to support rapid iteration.
Package prototypes as notebooks, APIs, or lightweight apps (FastAPI/Flask, Streamlit/Gradio) for internal and partner testing.
Define and track MVP metrics (e.g., lift vs. baseline, MAPE/MAE, precision/recall) and clearly communicate trade‑offs.
Partner on A/B or shadow tests and document assumptions, data gaps, and next steps; handoff stable work to engineering for hardening.
Collaborate closely with Engineering to scale MVP models to Production-ready.
Manage the data engineering infrastructure (warehousing/lake, pipelines, observability, schemas) in partnership with Engineering.

🤔 Who you are / what we're looking for

Strong Python (pandas/NumPy/scikit‑learn/polars) and SQL; familiarity with XGBoost/LightGBM; bonus for PyTorch/TF.
Comfort with time‑series and forecasting (statsmodels/prophet/gradient boosting; deep learning familiarity a plus).
Experience with anomaly detection, evaluation under noisy/drifting data, and practical feature engineering.
Pragmatic Data Engineering and MLOps: Git, Docker, basic cloud proficiency (GCP/AWS/Azure), basic CI/CD awareness, basic orchestration awareness (k8s, ECS/EKS), ETL and job management (dagster, airflow, dbt), experiment tracking (MLflow/W&B),
MLOps/DevOps experience (CI/CD for ML, IaC like Terraform, Docker/Kubernetes, model serving/monitoring, observability).
Data Warehouse/Lake management (dimensional modeling & ELT, performance tuning and cost control, governance & quality, batch/stream ingestion, lakehouse patterns). Tools: BigQuery/Snowflake/Redshift/Databricks, dbt, Airflow/Prefect/Dagster, S3/GCS/ADLS, Parquet/ORC + Iceberg/Delta/Hudi, catalog/lineage (Glue/Hive/Unity, OpenLineage), data quality (Great Expectations/Monte Carlo).

Nice to have:

Renewable energy context is a plus (PV/wind forecasting, weather/irradiance data, SCADA/IEC protocols, ISO/RTO datasets).
Geospatial tooling (PostGIS, geopandas, shapely) and weather/solar resource datasets.
Optimization (linear/mixed‑integer) for scheduling/dispatch; feature stores; stream processing (Kafka, Kinesis, pub/sub).

💓 What we offer

4 day work weeks
100% remote and flexible work environment
Competitive base salary
Opportunity to shape the engineering culture of a high-impact startup

💓 Our Shared Values

Focused on Outcomes: We build a relevant, simple and usable solutions. Constantly understanding, testing and improving.
Extreme Ownership: We are autonomous, ambitious and proactive and results focused. We value freedom and accountability.
Data Driven: Our decisions are powered by data: We use data to enhance our customers operations.
People-first: We prioritize well-being, empathy and purposeful work. We connect, recharge and deliver.