Technical Project Manager

We need to output HTML fragment with sections for headings present in posting: About the role? Actually the posting includes "About LILT AI..." then "Technical Project Manager — Applied AI". Then sections: Key Responsibilities (with sub-bullets), Essential Qualifications, Preferred Qualifications, Our Story, Our Tech, LILT in the News, etc. We need to use

for section headings like About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. However we should only include sections when posting has real specific content. The posting does not have explicit "About the role" heading but we have a paragraph describing the role after title. Could treat that as "About the role". The instructions: Use

for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). Do NOT repeat the job title as a heading. So we can create headings: About the role, Responsibilities, Requirements (maybe combine Essential and Preferred Qualifications?), Qualifications, Skills? But we need to map content. We have Essential Qualifications and Preferred Qualifications. Could combine into a single Qualifications section with two sublists? Or separate? The instruction: prefer a few substantive sections over many thin ones. So maybe we have sections: About the role, Responsibilities, Qualifications (including Essential and Preferred), Our Story (maybe not needed as not in list of allowed headings). The allowed headings list is examples, not strict? It says use

for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). It implies we should use those headings when applicable. Other headings like "Our Story" maybe not appropriate; we could omit or incorporate into About the role? Probably we should only include sections that match those headings. The posting does not have Benefits, Pay, Schedule. So we omit those. Thus we will have:

About the role

with paragraph describing role.

Responsibilities

with

• items from Key Responsibilities (including sub-bullets). Need to preserve all substantive information; we can flatten bullet points.

  Qualifications

  with
  • for Essential Qualifications and Preferred Qualifications maybe combined. We could also have

    Skills

    but not needed. We must not include boilerplate like EEO, etc. We need to remove duplicated lines: In Responsibilities there is a duplicated line "Manage Applied AI data collection and evaluation projects from technical scoping through validation, delivery, and retrospective review. through implementation, validation, delivery, and retrospective review." It seems duplicated; we should keep only one instance. Also there are duplicated sentences like "Coordinate requirements for data formats, schemas, metadata, annotation tools, and system integrations." appears twice; we should keep one. We need to preserve all substantive information, but remove duplicates. Let's extract content. First, About the role paragraph: After title "Technical Project Manager — Applied AI" we have: "We are looking for a Technical Project Manager to join our Applied AI team and lead large-scale multilingual data collection, human-in-the-loop workflows, and Large Language Model (LLM) evaluation initiatives. You will work with AI specialists, engineering teams, specialized Subject Matter Experts (SMEs), and data operations teams to translate technical AI requirements into high-throughput data workflows, hybrid evaluation pipelines, and measurable delivery plans. This role combines technical project management with hands-on data analysis, unit economics optimization, and a strong operational understanding of modern LLM training and evaluation processes." That's the description. Now Responsibilities: Under "Key Responsibilities" there are subheadings: Technical Project Delivery, Data Workflows & Evaluation, Quality, Cost, & Performance, Stakeholder & Contributor Management. We need to capture bullet points under each. Let's list them: Technical Project Delivery: - Manage Applied AI data collection and evaluation projects from technical scoping through validation, delivery, and retrospective review. (duplicate line appears) - Translate technical AI requirements into structured data specifications, domain-specific annotation requirements, evaluation plans, and acceptance criteria. - Manage timelines and throughput velocity, dependencies, delivery risks, and technical blockers across Applied AI, engineering, and operations stakeholders. - Coordinate data collection and annotation for supervised fine-tuning (SFT), preference data used in reinforcement learning from human feedback (RLHF), and model evaluation. Data Workflows & Evaluation: - Partner with engineering teams to define and validate data workflows, including ingestion, annotation tooling, systems integrations, quality checks, data generation pipelines, and delivery. - Operationalize hybrid evaluation methodologies, combining human review with automated evaluation tooling, and LLM-as-a-judge frameworks. - Coordinate requirements for data formats, schemas, metadata, annotation tools, and system integrations. - Coordinate requirements for data formats, schemas, metadata, annotation tooling, and automated quality checks. (duplicate) - Work with Applied AI specialists to operationalize evaluation methods, including human review, response ranking, and safety testing. - Investigate data, pipeline, or tooling issues and coordinate prompt resolution with technical owners. Quality, Cost, & Performance: - Use SQL and business intelligence tools to analyze throughput velocity, quality, unit economics, and supplier/delivery performance. - Define and monitor quality standards, including inter-annotator (IAA) agreement, gold-set performance, and dataset completeness. - Lead root-cause analysis on quality discrepancies, workflow bottlenecks, benchmark discrepancies, and quality edge cases. - Establish delivery validation checks and coordinate stakeholder acceptance of completed, accurate datasets, delivery targets, and evaluation results. Stakeholder & Contributor Management: - Translate complex technical requirements into actionable plans and clear instructions for global contributor teams, external vendors, and specialized domain SMEs and teams. - Communicate project status, technical risks and tradeoffs, unit economics, and operational risks to technical and non-technical stakeholders. - Ensure contributor feedback and evaluation findings inform improvements to internal tooling, guidelines, and dataset pipelines. Now Qualifications: Essential Qualifications: - 3–5+ years of technical project management experience in AI/ML, data platforms, or technical data operations. - Strong understanding of LLM training and evaluation concepts, including SFT, RLHF, human evaluation, automated evaluations, LLM-as-a-judge, and red teaming. - Proficiency in SQL for analyzing delivery velocity, data quality metrics, and cost structures. - Working knowledge of data pipelines, structured data formats, APIs, and validation processes. - Experience partnering with engineers to manage technical requirements, dependencies, and issue resolution. - Proven ability to track and optimize project unit economics (cost per token/task) and delivery velocity. - Strong communication skills, including the ability to translate technical requirements for multilingual audiences and diverse global contributor networks - Experience managing complex workflows using Agile, Scrum, or Kanban. Preferred Qualifications: - Experience using Python or scripting tools for data analysis, workflow automation, or evaluation scripting. - Experience managing high-skill Subject Matter Experts (SMEs) or specialized contributor networks across technical domains. - Experience with annotation platforms, automated evaluation tools, business intelligence tools, and Jira. - Experience delivering complex multilingual or multimodal data projects. - Background in computer science, data science, engineering, or equivalent practical experience. - Fluency in an additional language. We could combine into Qualifications section with two subheadings? But we only allowed

    for section headings; we could still have

    inside? Not specified. Safer to just list all qualifications in one list, maybe separate Essential and Preferred as separate

    with preceding

    indicating essential vs preferred. But we must not invent headings not in list. Could use

    to label. Thus Qualifications section:

    Qualifications

    then

    Essential Qualifications:

    then
    ... then

    Preferred Qualifications:

    then
    ... Now we must not include sections like Our Story, Our Tech, LILT in the News, etc. Those are not in allowed headings list, but they are substantive. However instruction: "Use

    for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule)." It doesn't forbid other headings, but says use

    for section headings (e.g. those). It might be okay to use other headings but they gave examples. Safer to stick to those headings only. The instruction also: "Prefer a few substantive sections over many thin ones." So we could omit Our Story etc. However we might lose substantive info. But maybe we can incorporate that info into About the role? The About the role currently only has the description paragraph; we could extend it to include company overview from "About LILT AI..." paragraph at top. Indeed the beginning includes "About LILT AI is changing how the world communicates — and LILT is leading that transformation..." That's substantive about company. We could include that in About the role section as well. Thus About the role section can contain both company overview and role description. Let's craft About the role paragraph(s): combine the initial "About LILT AI..." paragraph and the role description. We must not repeat job title as heading. Now produce HTML:

    About the role

    LILT is leading the transformation of global communication by making the world's information accessible regardless of language, using cutting‑edge AI, machine translation, and human‑in‑the‑loop expertise to translate content faster, more accurately, and more cost‑effectively without compromising brand, voice, or quality. Trusted by Intel, Canva, the U.S. Department of Defense, the U.S. Air Force, ASICS, and hundreds of global enterprises, and backed by Sequoia, Intel Capital, and Redpoint, LILT is building a category‑defining company in a $50B+ translation market being redefined by AI.

    We are looking for a Technical Project Manager to join our Applied AI team and lead large‑scale multilingual data collection, human‑in‑the‑loop workflows, and Large Language Model (LLM) evaluation initiatives. You will work with AI specialists, engineering teams, specialized Subject Matter Experts (SMEs), and data operations teams to translate technical AI requirements into high‑throughput data workflows, hybrid evaluation pipelines, and measurable delivery plans. This role combines technical project management with hands‑on data analysis, unit economics optimization, and a strong operational understanding of modern LLM training and evaluation processes.

    Then Responsibilities. We need to ensure we removed duplicate lines. Let's craft Responsibilities
    with each bullet as
    • . We'll combine similar points. List bullets: - Manage Applied AI data collection and evaluation projects from technical scoping through validation, delivery, and retrospective review. - Translate technical AI requirements into structured data specifications, domain‑specific annotation requirements, evaluation plans, and acceptance criteria. - Manage timelines and throughput velocity, dependencies, delivery risks, and technical blockers across Applied AI, engineering, and operations stakeholders. - Coordinate data collection and annotation for supervised fine‑tuning (SFT), preference data used in reinforcement learning from human feedback (RLHF), and model evaluation. - Partner with engineering teams to define and validate data workflows, including ingestion, annotation tooling, systems integrations, quality checks, data generation pipelines, and delivery. - Operationalize hybrid evaluation methodologies, combining human review with automated evaluation tooling, and LLM‑as‑a‑judge frameworks. - Coordinate requirements for data formats, schemas, metadata, annotation tools, and system integrations. - Work with Applied AI specialists to operationalize evaluation methods, including human review, response ranking, and safety testing. - Investigate data, pipeline, or tooling issues and coordinate prompt resolution with technical owners. - Use SQL and business intelligence tools to analyze throughput velocity, quality, unit economics, and supplier/delivery performance. - Define and monitor quality standards, including inter‑annotator (IAA) agreement, gold