About Basic AI
BaseSimple AI is building the data and benchmark infrastructure for the next generation of voice AI. We help AI understand people of all languages and accents.
The founders are ex-Meta product and engineering leaders from MIT and Brown University. We are a small, highly-owned team working directly with frontier AI labs to advance cutting-edge technology on audio models.
we are looking for one Strategic Projects Lead – Audio Data Owning high-priority audio data projects from start to finish through our platform.
about the role
This is an extremely high-ownership role at the intersection of Strategic operations, audio data, AI data delivery, product and customer execution.
You will have complex audio collection and annotation projects from customer requirement to final delivery. You’ll translate vague customer requirements into executable workflows inside our platform, run pilots, manage contributors and reviewers, track quality and throughput, identify bottlenecks, and ensure the final dataset meets customer expectations.
Because our platform is still evolving, this role is not simply to drive existing workflows. You will also identify gaps in the platform, define product requirements, and work with engineering to build or improve features needed to successfully deliver projects.
This is not a typical project management role. We are looking for someone who has personally driven disorganized, cross-functional projects from zero to completion, ideally in AI data, data labeling, annotation, localization, or crowdsourced operations.
what will you do
- Own audio data collection and annotation projects from kickoff to final customer delivery.
- Translate customer requirements into project specifications, contributor workflows, annotation guidelines, QA rubrics, acceptance criteria, and delivery plans.
- Configure and operate projects through Basesimple’s internal platform.
- Design and run a pilot to validate work design, contributor fit, audio quality, tooling, throughput, cost, and QA process before scaling.
- Manage day-to-day execution across contributors, annotators, reviewers, QA leads, and internal tools.
- Monitor project health based on quantity, quality, rejection rate, rework rate, cost, margin and timeline risk.
- Identify platform gaps that prevent projects from scaling, then write clear product requirements or feature requests.
- Partner with engineering/product to build or improve tools for project setup, contributor workflow, QA, reviews, payments, reporting, and delivery.
- Partner with contributor development to ensure we have the right supply based on language, accent, demographic, device, skill set, or job type.
- Create dashboards, trackers, and operating cadences for project execution.
- Clearly communicate project status, risks, tradeoffs, and blockers to founders, internal teams, and customers.
- Create repeatable playbooks for future audio collection, transcription, annotation, and QA projects.
- Run root-cause analysis when projects miss expectations for quality, cost, or timelines.
what we are looking for
- 3-7+ years of experience in data operations, AI data delivery, annotation operations, localization project management, marketplace operations, program management, or similar roles.
- Proven experience of ownership of projects from start to finish, from vague requirements to final delivery.
- Strong operator mindset: You can break down vague goals, make a plan, execute quickly, and unblock yourself.
- Experience managing complex workflows involving distributed contributors, reviewers, contractors, vendors, or large-scale data operations.
- Strong product understanding; Able to identify when tooling or platform features are needed and translate operational pain points into clear product requirements.
- Strong analytical abilities; Comfortable with spreadsheets, dashboards, funnel metrics, QA metrics, and operational KPIs.
- Excellent written communication; Able to write clear instructions, guidelines, SOPs, customer updates and internal product descriptions.
- Strong quality judgment and attention to detail.
- Comfortable balancing quality, speed, cost, customer needs, contributor experience, and platform constraints.
- It’s comfortable to work in ambiguity and redesign processes.
- High ownership, low ego, and willingness to deal with messy operational details.
strong pluses
- Experience with a data labeling, AI data, localization, or crowdsourcing company such as Scale AI, Surge AI, Appen, Telus Digital, RWS, TransPerfect Dataforce, Velolocalize, Lilt, Turing, DataAnnotation, Outlier, Remotetask, or similar.
- Experience with end-to-end delivery of data collection, annotation, transcription, evaluation, or QA projects.
- Experience with audio, speech, voice, ASR, TTS, speech-to-speech, transcription, podcast/audio production, or linguistic data.
- Experience building or improving internal tools, workflow systems, annotation platforms, QA systems, or contributor-facing products.
- Experience designing annotation guidelines, QA rubrics, reviewer training, or calibration workflows.
- Experience with multilingual or locale-specific data projects.
- Experience managing large distributed teams of contributors, reviewers, contractors or vendors.
- Basic SQL, Python, Airtable, Retool, no-code automation, or workflow tooling experience.
Example Projects You May Own
- Launching a new audio collection project in the priority language through our platform.
- Designing contributor workflows for voice actor auditions, recording tasks, metadata collection, and QA review.
- Aggregating 1,000+ hours of natural speech from contributors in a specific language or location.
- Building operating procedures to detect low quality audio, incorrect location, synthetic voice, background noise, or incomplete metadata.
- Running transcription or annotation workflows for speech datasets.
- Identifying whether our platform needs a new QA queue, reviewer dashboard, contributor instruction flow, or reporting feature – then writing the requirements and working with engineering to ship it.
- Running a pilot, diagnosing quality issues, improving workflow, and scaling the project to full production.
- Creating a reusable playbook that allows future customer projects to launch faster and with fewer manual steps.
what success looks like
You will be successful if Basesimple can repeatedly deliver audio datasets:
- accepted by customers
- delivered on time
- within budget
- High enough quality for model training
- Operationally repeatable through our platform
- Improve cost, quality, throughput and automation over time
Your key metrics may include:
- Approved audio hours delivered
- on time delivery rate
- customer acceptance rate
- Rejection/Rework Rate
- cost per approved hour
- Reviewer Agreement/QA Stability
- contributor throughput
- project margin
- Platform issues identified and resolved
- Number of reusable workflows or features created
Why join us?
This is a high-impact role at an early-stage AI company. You won’t just manage projects – you’ll help create the operating system for how to produce high quality audio data at scale.
You’ll work directly with the founders, own customer-critical projects, shape our internal platforms, and define the playbooks we use to deliver audio data for Frontier Voice AI models.
<a href