Innodata Inc. · via Innodata

Senior Data Annotator

Hybrid - Washington D.C

<div class="content-intro">Innodata (Nasdaq: INOD) is a global data engineering company. We believe that data and Artificial Intelligence (AI) are inextricably linked. Our mission is to enable the responsible advancement of artificial intelligence by providing the data, evaluation frameworks, and human expertise required to build AI systems that can be trusted at scale. We provide a range of transferable solutions, platforms, and services for Generative AI / AI builders and adopters. In every relationship, we honor our 36+ year legacy delivering the highest quality data and outstanding outcomes for our customers.</div>About this Role: Own annotation quality for a delivery team. A Senior Annotator personally handles the hardest annotation work, runs the QA process everything else passes through, trains and develops junior annotators, and serves as the final check before a dataset goes to the customer. <h3>Key Responsibilities:</h3> <h3>Quality Assurance:</h3> <ul> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Execute the review and audit plan for each batch, including which work gets sampled at what rate and when a batch gets sent back.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Perform frame-level and sequence-level review against the ontology and acceptance criteria; adjudicate disputed labels and issue a binding call.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Monitor platform quality reporting and annotator performance data, and act on what it surfaces.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Maintain gold-standard and benchmark task sets used to check individual and team accuracy.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Run consensus reviews on overlapping assignments and drive the guideline changes that close recurring disagreement.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Manage the rework loop: route corrections, verify closure, and confirm the fix generalizes across the batch, not just the sampled frames.</li> </ul> <h3>Training Junior Annotators:</h3> <ul> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Onboard new annotators onto the ontology, tooling, and workflow, and certify them against a benchmark set before releasing them onto production work.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Build and maintain training materials: worked examples, edge-case libraries, and short reference guides for each project.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Run calibration sessions where the team works the same difficult frames and reconciles differences against a documented standard.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Coach individuals using specific examples from their own reviewed work, and confirm the correction holds on the next batch.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Identify who is ready for harder task types or a new modality, and who needs additional support.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Serve as the first escalation point for annotator questions, answering in a way that turns a one-off answer into a documented rule.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Help screen candidates through practical annotation assessments.</li> </ul> <h3>Delivery and guidelines:</h3> <ul> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Serve as the quality gate for customer deliveries and provide the quality summary that accompanies each one.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Author and version labeling guidelines, decision trees, and edge-case libraries derived from the customer ontology.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Consolidate annotator questions into ontology clarification requests and route them through program management.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Perform the hardest annotation work personally: dense scenes, long tracking sequences, degraded or ambiguous imagery.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Surface quality, staffing, and capacity risk to program management early enough to act on.</li> </ul> <h3>Must-Have Qualifications: </h3> <ul> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">3+ years of image and video annotation experience, including 1+ year in a QA, review, lead, or adjudication capacity.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Demonstrated ability to define quality criteria and review against them — not only to label well personally.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Experience training or mentoring annotators, with evidence of measurable improvement in their output.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Solid grasp of what annotation quality metrics indicate and how to act on them.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Experience writing labeling guidelines or SOPs that other people worked from.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Ability to communicate quality risk clearly and early to program management.</li> </ul> Nice-to-Have Qualifications: <ul> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Experience across multiple domains — aerial, overhead, automotive, industrial, medical, or defense imagery.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Experience with synthetic data and the failure modes that distinguish it from real sensor data.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Experience with model-assisted labeling pipelines.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Experience supporting milestone-based deliveries with defined acceptance criteria.</li> <li style="font-size: 12pt; font-family: helvetica, arial, sans-serif;">Prior government or regulated-industry program experience.</li> </ul>   Program eligibility:  Some programs require eligibility for a government background investigation or credentialing. Assignment to those programs is contingent on meeting those requirements.<div class="content-conclusion"> Please be aware of recruitment scams involving individuals or organizations falsely claiming to represent employers. Innodata will never ask for payment, banking details, or sensitive personal information during the application process. To learn more on how to recognize job scams, please visit the Federal Trade Commission’s guide at <a href="https://consumer.ftc.gov/articles/job-scams." target="_blank">https://consumer.ftc.gov/articles/job-scams.</a>  If you believe you’ve been targeted by a recruitment scam, please report it to Innodata at <a href="mailto:verifyjoboffer@innodata.com" target="_blank">verifyjoboffer@innodata.com</a> and consider reporting it to the FTC at <a href="http://reportfraud.ftc.gov/" target="_blank">ReportFraud.ftc.gov</a>. </div>

Apply on Innodata ↗

Listing sourced from Innodata. Annotation Academy is independent of these platforms and does not guarantee work or pay. See our disclosures.