Founding ML Engineer

Open Select; San Francisco (CA); On-site; USD 150,000 - 300,000; Full time

Benefits offered by this job

Salary range $150K to $300K

Equity at an early stage

Direct path toward founding opportunities

Job summary

Open Select, an early-stage AI data company in San Francisco, is seeking a Machine Learning Engineer. You will be responsible for owning ML systems that convert raw data into structured intelligence.

The ideal candidate will have at least 3 years of experience in shipping ML models, particularly focusing on NLP. You will work on innovative retrieval systems and contribute to a team that has already achieved $7M ARR.

Qualifications

• 3+ years shipping ML models in production involving NLP or information retrieval.
• Experience in training and fine-tuning transformer encoder models.
• Ability to write strong Python and PyTorch code.

Responsibilities

• Build and ship retrieval systems to match job titles across languages.
• Resolve duplicate company records across multiple sources.
• Map raw job titles to canonical titles and functions.

Skills

Machine Learning (ML)

Natural Language Processing (NLP)

Python

PyTorch

Information Retrieval

Job description

An early-stage AI data company that went from zero to $7M ARR in its first 12 months. Their APIs power hundreds of customers by indexing hundreds of millions of professional profiles and company records from across the web. You will own the ML systems that turn that raw, multilingual, web-scale data into structured intelligence.

What you\'ll do

• Build and ship retrieval systems that match semantically equivalent job titles across languages.
• Resolve duplicate company records automatically across millions of entries from multiple sources.
• Infer organisational structures from raw people data, including reporting lines and team shapes.
• Detect technology usage signals from unstructured, web-sourced text at scale.
• Classify job transitions as promotions, lateral moves, demotions, or title edits.
• Map raw job titles to canonical titles, seniority levels, and functions across dozens of languages.
• Train and fine-tune encoder models from research paper through to production deployment.

Who you are

• Bring 3+ years shipping ML models in production, covering NLP or information retrieval.
• Train and fine-tune transformer encoder models, not just call third‑party APIs.
• Build, evaluate, and iterate on retrieval systems, classifiers, and embedding models.
• Use LLMs for structured extraction, classification, or data generation at scale.
• Write strong Python and PyTorch code with production quality.
• Work with a founder mentality, either as a past founder or someone building toward it.
• Bonus: Published research or open‑source contributions in NLP or information retrieval.

Python, PyTorch, NLP, LLMs, transformer architectures, embedding models, information retrieval, entity resolution, text classification, contrastive learning, metric learning, representation learning, LLM inference pipelines, distributed training frameworks

Why you\'ll thrive

• Earn $150K to $300K, with equity at an early stage before significant dilution.
• Own the full ML research and engineering cycle, from prototype to production.
• Solve genuinely hard problems on a dataset of hundreds of millions of real‑world records.
• Join a small team that has already proven commercial traction with $7M ARR in year one.
• Work in a role with a direct path toward founding something of your own.