Effective AI - Data Product Engineer

• →5 to 8 years of experience in data engineering, building and operating production data pipelines and systems [Required]

• Built and scaled a data system end-to-end: connected new external data sources, owned ingestion through production [Must have]
• Recent experience at high-talent-density companies or startups (Seed to Series D, strong bigtech, AI-native companies, fast-moving fintech) [Required]
• Built agent harnesses or LLM-powered extraction/validation workflows [Strongly preferred]

• CS (or STEM) degree from a top-tier university [Required]

• Production data pipeline design, ingestion, and orchestration [Must have]
• Experience with AI/ML agent frameworks and eval harnesses [Required]
• Large-scale unstructured document processing (PDFs, filings) [Strongly preferred]

• Based in SF or willing to relocate; in-office 5 days/week [Must have]
• Authorized to work in the US (H-1B transfer, TN, or citizen/GC) [Must have]

• Pure ML/data science profile with no data engineering or pipeline ownership
• Prefers large-company structure and slow iteration cycles

• Salary | $230K-$280K (posted); intake stated $210K-$270K, flex to $290K
• Equity | Competitive equity
• On-site policy | 5 days in-office in San Francisco, CA
• Visa sponsorship | Open to visa transfers (OPT, H-1B transfers); US work auth required
• Employment type | Full-time
• Location | San Francisco, CA

1. Are you able to work in San Francisco and come into the office 5 days per week?
2. Describe a time you connected or ingested a new external data source into a product. What was the source and how did you make it usable?
3. What's the most complex data project you've built end-to-end? Walk us through what made it hard and how you scaled it.
4. Can you be on-site? If not, are you willing to relocate?
5. What is your salary expectation?
6. How actively are you exploring new opportunities?

Updated Jul 29, 2026

• Traditional insurance carriers (no startup DNA, not technically challenging): Farmers, Nationwide, State Farm, Allstate, Progressive, Liberty Mutual, GEICO, USAA, Travelers
• Already thoroughly sourced or off-limits per HM: Rubrik, Nirvana Insurance

Note: "Data infrastructure and pipeline companies" listed 9 of 10; 1 company was not expanded before copy and is missing.

For reference only, do not source these specific profiles.

Muhammad Janjua - LinkedIn Data Engineer at Meta | High-throughput data pipelines & cloud infra | San Francisco Bay Area

• Strong communication; explained ad-campaign pipeline complexity well
• Decent day-to-day agent experience; owns data pipelines across multiple teams
• Weak spots: limited progress on technical task, couldn't explain modeling setup, pipeline explanation skewed business over technical

David Lyon - LinkedIn Software Engineer @ Meta | Data Science, ML, Python | Newark, US

• Feature engineering + SFT to detect bot farms; some pipeline building
• Weak spots: limited agentic experience, low energy, doubts on seed-stage velocity

Vivek Jain - LinkedIn Staff Software Engineer at Databricks | Palo Alto, US

• HM (Arijit) to share more detail during intake

Chetas Joshi - LinkedIn Data & AI @ Robinhood | San Francisco, US

• Great schools + companies; worked with Arijit at Rubrik; great feedback (not currently looking)

Note: "Show all 5 candidates" showed 4 of 5; 1 profile missing (not expanded before copy).

• Ownership in Production: Prioritize candidates who have built and operated fully productionized data pipelines with clear failure management and quality monitoring.
• AI/LLM Expertise: Require hands-on, recent experience with AI agents, RAG pipelines, and evaluation harnesses, not just traditional data engineering.
• Startup & High-Talent Background: Focus on seed to Series D or high-talent tech firms whose experience maps to the end-to-end nature of the role.
• Top-Tier Academic/Employer Signal: Strong emphasis on a CS or equivalent STEM degree from top-tier schools and companies; non-CS backgrounds require equally impressive top-tier signals.
• Specific rejection (Jul 30, 2026): One candidate rejected at HM Review, "does not meet our bar on school and employer."