Please mention DailyRemote when applying
Employment Type: Full-time, exempt
Work Arrangement: Remote
Location: US
Reports To: Chief Product Officer
Own the product roadmap for iLumos.ai
Prioritize features based on client ROI, competitive gaps, and engineering feasibility
Define quarterly OKRs and report delivery progress to leadership using AI-generated metrics
Identify underserved segments — boutique firms, solo practitioners, mid-market in-house teams priced out of premium platforms
Drive API-first and integration strategy
Write system prompts, tool definitions, and structured output specs for every LLM workflow
Design agentic, multi-step workflow architecture for patent analysis pipelines
Define evaluation rubrics for LLM-as-judge pipelines; LLM Ops wires them into CI/CD
Iterate on prompts using Braintrust eval data, attorney feedback, and hallucination reports — prompt changes ship like code
Maintain a versioned prompt library with eval scores and rollback capability
Serve as the final go/no-go quality gate on all AI-generated outputs before they reach clients
Validate output accuracy against professional standards expected by Am Law firms and Fortune 500 IP teams
Curate the gold evaluation dataset — attorney-reviewed ground truth used to benchmark every workflow
Liaise directly with law firm and corporate IP clients to convert feedback into prompt and pipeline improvements
Flag and escalate legal accuracy risks; one hallucinated citation in active litigation can cost a client millions
Run async-first, output-oriented continuous delivery — measured by what ships
Replace standups with AI-generated daily digests from commits and deployment events
Coordinate across Full-Stack and Platform/Security engineering on dependencies without adding process overhead
Own the enterprise client relationship: demo new capabilities, manage expectations, translate feedback into engineering constraints
Manage budget, vendor relationships (Anthropic, AWS, data providers), and toolchain discipline
Claim structure and construction (independent/dependent claims, means-plus-function, PHOSITA standards)
Claim charting and "reads on" infringement/invalidity mapping
Prior art analysis under §102/§103, including motivation-to-combine arguments
Office action review and prosecution history/estoppel strategy
Invalidity contentions and §112 written description/enablement issues
Patent portfolio strategy (citation analysis, continuations, FTO)
USPTO Patent Center, EPO OPS, WIPO PCT, and prior art tools (Derwent, Orbit)
LLM architecture fundamentals: context windows, temperature, sampling, and their impact on legal output quality
Prompt engineering: chain-of-thought, few-shot examples, structured JSON schemas, hallucination-prevention constraints
Tool use / function calling for patent database queries, prior art retrieval, and citation lookup
Multi-step agent design (LangGraph or equivalent) for stateful patent analysis pipelines
LLM evaluation: golden datasets, LLM-as-judge patterns, legal accuracy metrics, Braintrust or equivalent
Hallucination mitigation in legal contexts: citation fabrication, claim scope inflation, jurisdictional drift
Cost/latency tradeoffs: prompt caching, token economics, batching at enterprise scale
Daily tools: Anthropic Claude API, LangSmith, LangGraph (or equivalent), GitHub, JIRA
Experience: 3–10 years combined across patent domain work and B2B SaaS product management, with a preference for 5+ years in B2B SaaS product management, including at least 2 years at a company competing in a domain with significant legal or regulatory accuracy requirements.
Patent Domain Credentials (at least one required)
USPTO registered patent agent (registration number required), or
JD with substantial patent prosecution or litigation experience
3+ years hands-on experience with claim charts, invalidity contentions, FTO analyses, or office action responses
Familiarity with at least one patent analytics platform (Patlytics, Derwent Innovation, Orbit Intelligence, PatSnap, or Google Patents Advanced)
Technical undergraduate degree preferred (engineering, chemistry, CS, or biology)
Direct experience working with patent attorneys, Am Law firms, or in-house IP teams
Product & AI Leadership
Track record shipping AI-powered features to enterprise legal or IP clients
Comfort operating without a Scrum Master or dedicated QA team — async-first, high-ownership delivery
Ability to write LLM workflow specs precise enough for engineers to implement without clarification sessions
Strong client-facing skills: able to demo a half-built product and field hard technical/legal questions
Ability to make technical tradeoff calls with engineering (latency vs. accuracy, caching vs. freshness, context window vs. cost)
Sharp prioritization instincts in resource-constrained, high-leverage environments
A rare combination of patent domain expert and AI practitioner
Comfortable owning outcomes with minimal process, at any career stage from early-career to seasoned senior
Someone who has worked with attorneys or IP teams and understands the real stakes of inaccurate legal AI output
A builder with taste — you write specs engineers can implement and hold the bar on what ships to clients
Collaborative and decisive; low-ego but confident enough to disagree with an attorney on claim scope and be right
The product and prompt architecture behind iLumos.ai's core LLM workflows
The gold evaluation dataset and LLM-as-judge framework that sets the professional accuracy bar for legal AI
A scalable, async-first delivery model that lets a lean team outperform larger, slower incumbents
Direct relationships with Am Law firms and Fortune 500 IP teams that shape the roadmap
The category definition for AI-native patent intelligence — built on accuracy, trust, and real domain depth
Compensation and Pay Transparency
The expected total annual compensation range for this role is $135,000–$275,000, which may include base compensation and other eligible incentive components. This range reflects national market data for the role and is intended as a general guide; actual pay will vary based on the candidate's location, cost of labor in that geographic market, patent domain credentials, AI/LLM production experience, and overall seniority.
For candidates in jurisdictions that require pay range disclosures, Lumenci will provide a good-faith compensation range specific to the candidate's location as part of the job posting or during the interview process, consistent with applicable law. Where required, the location-specific range may differ from the range stated above.
Salary History: In compliance with applicable laws, Lumenci does not request, require, or rely on an applicant's prior salary history when determining a starting salary or during any part of the hiring process.
Stop the endless job search. Our AI finds and applies to the best jobs for you.
Discover remote opportunities in Software Development
Answer easy questions
200,000+ jobs across 15+ categories
Get your best job matches
Only hand-screened, legit jobs
Find a remote job faster
No ads, scams, or junk
“ I was the first applicant for a remote marketing position that got listed on the company website the same day I applied. Had an interview within 48 hours!