AI-Native Data Associate - Energy Data

Parisi Labs; Boston (MA); On-site; USD 65,000 - 85,000; Full time

Job summary

Parisi Labs builds world models for physical industry and is hiring an early-career data professional to own data ingestion and cleaning flows. You will source, ingest, clean, and reconcile data at scale, using agents to explore new sources and ensure provenance.

Expect hands-on work with Python, SQL, and agent-enabled pipelines in a startup-like environment. The role is on-site in NYC or Boston/Cambridge (about 3 days/week) and requires US work authorization.

Qualifications

• Strong data-quality judgment and curiosity about data anomalies.
• Experience building or running agent-driven data workflows (extraction, scraping, classification, entity matching).
• Proficiency in Python and SQL for reading, writing, and debugging pipeline code.

Responsibilities

• Ingest data at scale using agents and AI-powered pipelines to structure data from filings, documents, PDFs, websites, APIs and spreadsheets.
• Clean, de-duplicate, match entities, and reconcile data across sources with automation-first approaches.
• Maintain and extend ingestion pipelines and data stores, adding new sources and fields while keeping systems healthy.
• Explore and discover by using agents to find new datasets and patterns worth acting on.
• Verify data provenance with checks so AI-generated data can be traced back to its source.
• You will read/write and debug pipeline code in Python/SQL, and work daily with tools like Claude Code, Cursor, or MCP-connected agents.

Skills

Python

SQL

Data Pipelines

Airflow/dbt

Agent-driven workflows

Data quality judgment

AI/LLM tooling

Tools

Claude Code

Cursor

MCP-connected agents

OpenAI API

Anthropic API

Job description

Parisi Labs builds world models for physical industry, starting with power and energy markets. Products like Ask The Grid run on data that's large, messy, scattered and constantly changing.

We're hiring an early-career data person who works natively with AI agents. You'll own the flow of data into our existing systems: sourcing it, ingesting it, cleaning it, keeping it accurate and extending it.

We aren't looking for someone who sometimes asks ChatGPT for help. We want someone whose default way of working is to direct agents: building and running agent-driven workflows that ingest, clean and reconcile data at scale, and using agents to explore and find new sources we didn't know existed.

Share one example of real work you completed with AI agents or LLMs: what you built, how you ran it and how you verified the result. A link, repo or short write-up is fine. Finalists complete a short paid practical exercise, and using AI is encouraged.

• Ingest at scale: Use agents and LLM-powered pipelines to pull data from filings, utility and ISO documents, PDFs, websites, APIs and spreadsheets, then structure it.
• Clean and reconcile: Normalize, de-duplicate, match entities and fix conflicts across sources, using automation instead of hand edits wherever possible.
• Maintain and extend our systems: Keep existing ingestion pipelines and data stores healthy, add new sources and fields, and fix what breaks.
• Explore and discover: Use agents to find new datasets and spot gaps, anomalies and patterns worth acting on.
• Verify everything: Build checks so AI-generated data can be traced back to its source and trusted. Speed is only useful if the data is right.
• You've built or run agent-driven or LLM-powered data workflows (for example extraction, scraping, classification or entity matching), and you can show how you checked the output.
• You're comfortable working daily in tools like Claude Code, Cursor, MCP-connected agents or the OpenAI/Anthropic APIs.
• Working Python and SQL. You can read, write and debug pipeline code even when an agent drafted it.
• Strong judgment about data quality: you notice when something looks wrong and you go find out why.
• High agency and comfort with ambiguity. You'd rather build the tool than repeat the task.
• Students, recent grads and early-career people are encouraged to apply.
• Interest in energy, power markets or physical infrastructure
• Experience with Airflow, dbt or similar orchestration tools
• Startup or self-directed project experience
• $65–85K salary (full-time) or $30–40/hr (part-time/contract)
• NYC or Boston/Cambridge, ~3 days/week onsite
• Must be authorized to work in the US