Joulent is a full-stack energy technology company built to deliver reliable, multi-gigawatt power at the speed and scale the AI era demands. Backed by Engine No. 1 and GE Vernova, our modular, Across-the-Meter™ approach integrates generation, storage, and advanced controls to power industrial hyperscalers.
The Energy Intelligence team turns data into decisions: load- and market-facing forecasting across short- to long-term horizons, built on a shared, agent-native platform spine.
🚀 The Role
This is a founding, 0→1 build- You’ll design and own the full data and ML spine the entire Energy Intelligence org runs on, from data engineering (ingestion, pipelines, feature store, quality and lineage) through the MLOps lifecycle (training, serving, monitoring, and the agent/model gateway). Our data is rich rather than petabyte-scale, so the hard problems are messiness, freshness, and trust, not raw volume. You move fast, ship to production, and treat AI agents as force-multipliers. If you want to build something real that helps power gigawatts of compute for the AI era, this is the seat.
🛠️ What You’ll Do
Builder Mode: architect the spine
- Build production data pipelines and ML processes from scratch. You define the frameworks, SLAs, and standards everyone else builds on.
- Own data ingestion end to end: market data (ISO/RTO feeds, LMPs, nodal price history), weather, interconnection-queue, and plant/asset telemetry- managing quality, lineage, and freshness (these inputs carry direct P&L and reliability consequences).
- Build the feature store and simulation data plumbing, prevent training/serving skew, and ship AutoML pipelines so engineers train and deploy without rebuilding the plumbing each time.
Operator Mode: make it reliable, fast, and cheap
- Own the ML lifecycle end to end: training orchestration, serving, registry, CI/CD, monitoring, and drift detection.
- Stand up high-throughput, low-latency serving for intra-hour, day-ahead, and seasonal forecasts, with metric-aware alerting.
- Define SLOs and dataset SLAs (freshness, quality, lineage), support make-vs-buy across data feeds and the MLOps stack, and balance the trade-offs across performance, cost, and reliability, making the calls explicit and defensible.
AI-Native Mode: agents as force-multipliers
- Own the agent platform and model gateway (orchestration, routing, caching, and batching), with token budgets, guardrails, and observability across the model and agent fleet.
- Set the eval gates that decide which models and agents ship. Back-tests and evals are first-class here, with P&L and reliability on the line.
🎯 What You’ll Bring
- AI-native by default. You work alongside coding and analysis agents as a force-multiplier, not a novelty, and you stay current as the tooling moves. You raise the standard for how the team builds with AI.
- High agency and urgency. You define requirements rather than wait for them. You know when the 60%-right-now answer beats 95% answer in 3 days.
- Cross-functional drive. You align engineers, data vendors, and external partners without needing authority.
🧰 The Technical Bar
- 5+ years in a full-stack data and ML platform or production engineering role, having built production data pipelines and ML serving.
- Expert SQL and strong Python, with software-engineering rigor: CI/CD, containers, and infrastructure-as-code.
- Hands-on experience running production data and ML pipelines end to end with orchestration tools like Airflow, across batch and streaming workloads on a modern cloud data platform.
- A track record of preventing training/serving skew, with data quality, lineage, and observability built into your pipelines.
- Experience operating model serving, experiment tracking, and drift monitoring in production.
- B.S. or M.S. in Computer Science, Machine Learning, Data/Software Engineering, or comparable field, or equivalent experience.
✨ Bonus Points
- Experience with energy/market data (ISO/RTO, LMPs, nodal pricing, weather, interconnection-queue feeds); time-series data at scale; graph data and embeddings.
- LLM and agent ops: model gateways and routing, prompt and eval frameworks, token-cost optimization.
- Production cost modeling and simulation (PLEXOS, Aurora) and inference efficiency (quantization, distillation, transfer learning) for latency and cost budgets.
💡 Why You’ll Love It Here
- Build something foundational. A true inflection moment in the company’s trajectory as we inch deeper in the energy supply chain and you get to build the data plumbing that makes it happen.
- Ownership and autonomy. Work with a fast growing, seasoned team where you own major parts of the system end to end, directly influencing the bottom line
- Scale that matters. A well-funded mission powering multi-gigawatt, national-scale infrastructure for AI-scale compute and industrial projects.
💰 Compensation
- Base salary range: $175,000–$230,000, depending on experience. Final offer will be based on the candidate’s skills, experience, and qualifications relative to the role’s requirements.