About SmartHelio
SmartHelio is a Y Combinator company (YC W22). We are a customer-focused and research-driven deep-tech startup headquartered in Lausanne, Switzerland. We are disrupting conventional energy operations and asset management by bringing AI-driven asset management workflow automation. We enable solar/clean energy companies with automated system performance analysis and decision-intelligence capabilities for smoother asset management and higher profitability. Learn more at www.smarthelio.com
Role 1: Data Science Intern
Role 2: Software Engineering Intern
Role 3:
Duration - 6 Months
Location - Noida, India (on-site)
Start - Immediately
How to Apply
Please fill the application form using the link - https://docs.google.com/forms/d/e/1FAIpQLSfFoFf8AaxIkNFJpwlCLyiOA5JpvQHSlbkwHvvNUMe_3aUITQ/viewform?usp=header
About the Data Science Intern Role (Role 1)
You will transform raw PV data into reliable insights and create models that power our AI-driven asset management workflows. Expect hands-on work across data wrangling, feature engineering, model development, evaluation, and reporting—plus close collaboration with our business and customer team.
What You Will do
Data wrangling & QA/QC:
Clean, align, and validate SCADA/CMS/IoT datasets; fix gaps, outliers, drifts, and timestamp issues.Feature engineering:
Build features from inverter/string/meter/weather streams for PR/CUF, availability, clipping, soiling, and curtailment analysis, and other advanced fault detections.Modelling:
Develop and validate ML/statistical/Agentic/LLM-based models for anomaly detection, fault validation, baseline/expected energy, and degradation trends.Agentic workflows
: Define guardrails and evaluation metrics for AI agents; benchmark precision/recall and write succinct (≤50 words) rationales for decisions.Analytics & reporting:
Produce daily/weekly loss waterfalls, incident timelines, and concise client-ready narratives. Assist in analyzing datasets to extract actionable insights, and create clear, visually engaging reports and dashboards that support data-driven decision-making.Tooling:
Package analyses in reproducible notebooks/scripts; contribute to lightweight MLOps (experiment tracking, data/version hygiene).
What You Will Need
- Education: Bachelor’s degree (completed or final year, considering you are available for a 6 months internship program) in Engineering/Science—Computer Science, Mathematics, Electrical, Renewable, Energy, or related.
- Solid Python (pandas, numpy), SQL, and MongoDB.
- Working knowledge of scikit-learn (bonus: PyTorch/TensorFlow), time-series methods.
- Basic understanding of Machine Learning and Deep Learning modeling
- Dashboards (Power BI/Tableau), Jupyter workflows, Git; basics of MLflow/DVC.
- Cloud data stacks (AWS/Azure), APIs, time-series stores.
- Nice to have - PVlib/PVsyst/SAM exposure; Meteonorm/SolarGIS/Solcast familiarity.
- Mindset: Detail-oriented, curious, and comfortable working with real-world messy data. Strong ownership and bias for action.
Location Requirement:
- You must be based in Noida and available for on-site work.
Compensation & Benefits
- A fair monthly stipend.
- Opportunity to join as a Full-time employee.
- Opportunity to grow in diverse emerging fields like AI, Data Science, and Product Engineering, etc.
About Software Engineering Intern Role (Role 2)
Join our platform team to build reliable data and application services that power AI agents and PV analytics. You will design and implement ingestion jobs, APIs, and microservices; harden reliability and observability; and help ship features used by asset managers daily.
What You Will do
Data ingestion & pipelines
: Build resilient connectors for SCADA/CMS/IoT feeds; implement validation, retry, and back-pressure.APIs & services:
Develop secure REST/GraphQL endpoints to serve KPIs, events, and reports; write clear specs and tests.Dashboards & UI:
Develop excellent visualization and application pages.Agentic integrations
: Expose tools/functions for AI agents; enforce guardrails, rate limits, and audit logging.Storage & performance:
Work with SQL/NoSQL/time-series stores (e.g., Postgres, MongoDB, Timestream/Influx/Timescale); profile and optimize queries.Cloud & DevEx
: Deploy on Azure (reading from AWS where needed); CI/CD, infra-as-code basics, monitoring/alerting, and runbooks.Quality:
Unit/integration tests, code reviews, and documentation to productionize workflows.
What You Will Need
- Education: Bachelor’s degree (completed or final year, considering you are available for a 6 months internship program) in Engineering/Science—Computer Science, IT, Electrical, Electronics, or Related
- Frontend: Frontend familiarity (comfortable with React).
- Backend: Python (FastAPI/Django) or TypeScript/Node.js (Express/NestJS).
- Data: SQL, schema design, pagination; working with time-series.
- Cloud & tooling: Basics of Docker, Git, CI; comfort with Linux CLI.
- Nice to have: Experience with Azure (App Service/AKS), AWS (S3/Lambda) or Observability systems (OpenTelemetry/Prometheus/Grafana), caching (Redis), and auth (OAuth2/JWT).
- Mindset: Detail-oriented, curious, and comfortable working with real-world messy data. Strong ownership and bias for action.
Location Requirement:
- You must be based in Noida and available for on-site work.
Compensation & Benefits
- A fair monthly stipend.
- Opportunity to join as a Full-time employee.
- Opportunity to grow in diverse emerging fields like AI, Data Science, and Product Engineering, etc.
About Solar Performance Engineering Intern Role (Role 3)
As a Solar Performance Engineering Intern, you will work closely with our Product/technology & Customer success teams in India and our research collaborators in Switzerland. You will help monitor, analyze, and improve the performance of utility-scale and C&I solar PV assets by turning raw data into actionable insights that drive the system's energy yield, availability, reliability, and profitability. Additionally, you will closely work with our AI engineers and product team to supervise, and review the AI Agents by defying agentic guardrails and performance benchmarks.
What You Will do
- Build and create the digital twin of the PV systems and successfully onboard PV plants.
- Analyse SCADA/CMS/plant data (inverter, string, meter, weather station) to verify and compute the KPIs (PR, CUF, availability, specific yield, etc.), track losses, and identify underperformance.
- Review and validate the daily faults and critical alarms.
- Review and validate the performance of AI Agents/Algorithms.
- Correlate time-series data, weather, and metadata to identify system underperformance root causes along with their severity.
- Draft/review/edit the system performance reports - summary, loss waterfalls, faults, issues, and incident timelines for asset managers and O&M team.
- Coordinate with client teams to gather requirements, align on KPIs, collect relevant data/information, and present findings with crisp explanations and recommended actions.
- Conceptualize the new algorithms in close collaboration with the Data Scientists.
- Develop models (mostly in python) to identify, validate, and review the system faults.
- Review the system meta data.
- Execute data quality checks (gaps, outliers, sensor drift, timestamp issues) and work with clients to resolve telemetry and metadata inconsistencies.
- Develop models to automate the data quality checks.
- Assist in building/validating baseline energy models, fault models, and customized algorithms.
- Develop custom models to conduct special analysis.
- Contribute to Python notebooks and lightweight scripts to standardize analyses
- Collaborate with engineers to productionize repeatable workflows.
What You Will Need
- Education: Bachelor’s degree (completed or final year, considering you are available for a 6 months internship program) in Engineering/Science—Electrical, Renewable, Energy, or related.
- Solid proficiency in Excel and intermediate Python for data wrangling/analysis.
- Strong electrical engineering fundamental (circuit theory, power system, basic inverter design, fundamentals of power electronics, etc.)
- Basic understanding of core PV concepts (irradiance, PR, CUF, DC/AC ratio, clipping, soiling, degradation, curtailment).
- Ability to analyze and comprehend large data sets using Python.
- Ability to generate data visualization using Python’
- Mindset: Detail-oriented, curious, and comfortable working with real-world messy data
- Strong ownership and bias for action.
Location Requirement:
- You must be based in Noida and available for on-site work.
Compensation & Benefits
- A fair monthly stipend.
- Opportunity to join as a Full-time employee.
- Opportunity to grow in diverse emerging fields like AI, Data Science, and Product Engineering, etc.
How to Apply
Please fill the application form using the link here (https://docs.google.com/forms/d/e/1FAIpQLSfFoFf8AaxIkNFJpwlCLyiOA5JpvQHSlbkwHvvNUMe_3aUITQ/viewform?usp=header).
Note:
