Spark (sparkhq.ai) is an AI-powered workflow platform for large-scale renewable energy projects. This includes community and utility-scale solar farms and battery storage projects. Tae Kim and Julia Wu were previously product and engineering leaders at Apple, Tesla, Lucid Motors, and Microsoft.

The heavy lifting in renewable energy projects is not in construction but in pre-construction paperwork. Over 2,000 energy-regulating entities, 3,000 counties, and 17,000 townships in the US have their own rules and permit applications. Project developers spend 5 to 50 hours analyzing the permitting, zoning, regulatory, and utility policies for every project location.

With Spark, developers can search across thousands of PDFs, maps, and websites to identify requirements and screen locations in seconds, not weeks.

Data aggregation, while very powerful, is only the beginning. Our workflow and document intelligence solutions follow projects from origination to financing to being operational.

Fresh out of YC W24, we are proud to have landed enterprise contracts and to serve some of the country's leading solar/storage development companies—some of which are subsidiaries of publicly traded companies.

We are proudly backed by YC, AI Grant (Nat Friedman / Daniel Gross), and other top-tier funds and angels (to be announced).

Read more: https://www.ycombinator.com/companies/spark

Our customers count on Spark to aggregate and organize the energy industry's data in real time and apply AI to it. In this founding role, you will work closely with our founders to build Spark’s data infrastructure, pipelines, and business logic from scratch. You will also be involved in customer interactions and support and help shape the product roadmap.

Our stack at a glance:

1. Data infrastructure: We index, crawl, scrape, and ingest thousands of websites daily. These pipelines use Typescript, Python, and OpenAI.
2. LLMs: We use OpenAI, Anthropic, Perplexity, and a number of chunking, retrieval, and RAG libraries for extraction, summarization, and question-answering. This is used on public and user-uploaded data.
3. Frontend: Typescript and React.js for our interactive web app.
4. Infra: We run on AWS, Vercel, and Porter.

While we proudly serve the energy vertical, Spark is, first and foremost, a tech company. This stems from our founders' backgrounds as backend, data infra, and full-stack software engineers.

Our customers count on us to aggregate and organize the data in this industry in real time and apply AI to it.

Our tech can be broken down into 3 components:

Data infrastructure: We index, crawl, scrape, and ingest thousands of websites daily. These pipelines use Typescript, Python, and OpenAI (Chat Completions API).

LLMs: We use OpenAI (Assistants API), Anthropic, Perplexity, and a number of chunking, retrieval, and RAG libraries for extraction, summarization, and question-answering. This is used on public and user-uploaded data.

Frontend: Typescript and React.js for our interactive web app.

We run on AWS, Vercel, and Porter.