Full-Time
Content Evaluator
Bilingual, Vietnamese and English
Innodata
1,001-5,000 employees
Delivers AI data engineering services
Compensation Overview
$20/hr
Remote in USA
Remote
 AI & Machine Learning |
|---|
Get referred to Innodata
See people who can refer or advise you
- Prior experience in customer service, call centers, retail, or handling customer communications via email, chat, or phone.
- Exceptional written English skills with a strong command of tone, brand voice, grammar, and nuance.
- Exceptional written Vietnamese skills with a strong command of tone, brand voice, grammar, and nuance.
- Ability to strictly follow multi-tier evaluation guidelines, complex logic trees, and technical rubrics without deviation.
- Comfort using dedicated web-based tools and labeling interfaces.
- Review and score AI-generated customer interactions across Foundational, Experiential, and Operational dimensions, including Action Fidelity, Faithfulness, Hallucination, Compliance, Tone, and Handoff.
- Benchmark informational (R1) and transactional (R2) customer queries against authoritative business sources, including FAQs, product catalogs, and standard operating procedures, within the task interface.
- Participate in dual-review processes and daily calibration audits to ensure inter-rater agreement and establish ground-truth performance targets.
- Deliver precise evaluations against performance targets.
- Prior customer service, call center, retail, or customer-communication experience is highly prioritized.
Innodata is a global data engineering company that provides AI-enabled software platforms and managed services to create high-quality training data and data pipelines for AI. It combines proprietary software with a global network of over 5,000 subject-matter experts to collect, create, annotate, and validate data, and it also offers synthetic data generation and end-to-end AI lifecycle services. The DDS segment drives most revenue, with Synodex and Agility supporting healthcare data and media monitoring. Its goal is to deliver reliable, safe, and effective datasets and AI lifecycle solutions across industries including technology, finance, insurance, and government.
Company Size
1,001-5,000
Company Stage
IPO
Headquarters
Hackensack, New Jersey
Founded
1988
Get referred to Innodata
See people who can refer or advise you
Simplify's Take
What believers are saying
- Q2 2026 revenue hit $92.1 million, up 58%, while guidance stayed above 40%.
- Rahul Singhal becomes CEO on September 30, 2026, preserving sales execution continuity.
- Customer concentration improved from 56% to 37% for the largest client, widening leverage.
What critics are saying
- Innodata filed a $300 million ATM on August 6, 2026; dilution can cap returns.
- The top two customers still generated 71% of Q2 2026 revenue, risking abrupt collapse.
- If frontier labs insource data engineering, Innodata’s revenue base can shrink within quarters.
What makes Innodata unique
- Innodata’s August 4, 2026 Cyber Training Suite turns security flaws into trainable benchmarks.
- Palantir selected Innodata on January 29, 2026 for multimodal AI data engineering.
- Q2 2026 showed 49% adjusted gross margin, proving premium AI dataset economics.
Help us improve and share your feedback! Did you find this helpful?
Benefits
Flexible Work Hours
Remote Work Options
Company News
Three AI-related stocks — Vertiv Holdings, Applied Optoelectronics, and Innodata — are trading well below their 2026 highs despite strong operational growth. Vertiv's second-quarter sales rose 24% to $3.27 billion, with adjusted operating margin expanding 410 basis points to 22.6%. The company sells power and cooling solutions for AI infrastructure. Insider Monkey counted 112 hedge funds holding Vertiv in Q2, up from 96 in Q1. Applied Optoelectronics posted record revenue of $191.9 million, with 800G volume more than doubling from Q1. Management expects 800G and 1.6-terabit demand to exceed production capacity through mid-2027. However, the company reported a $22.8 million GAAP loss. Innodata's revenue grew 58% to $92.1 million, with adjusted EBITDA rising 92% to $25.4 million. The company provides data engineering and model-evaluation services for AI applications. Each stock faces distinct risks around timing, customer concentration, and operational execution.
Innodata has launched its AI Cyber Training Suite, a cybersecurity product targeting AI coding agents and enterprise software security. The suite introduces a methodology for training and evaluating AI agents on software vulnerability detection and patching. The product addresses concerns about trust, code quality, and security when deploying AI agents in large software environments. It targets enterprises relying on AI-generated code and legacy system modernization. The stock closed at $62.33, up 17.6% year to date and 43.2% over the past year, though down 9.7% over the past month. The launch reinforces Innodata's strategy of building proprietary tools that integrate into model builders' workflows. However, the company faces challenges from concentrated client reliance and competition from larger security vendors like CrowdStrike.
Innodata reported second-quarter 2026 revenues of $92.1 million, up 58% year-over-year, beating the Zacks Consensus Estimate of $86.3 million. Earnings of $0.41 per share exceeded the $0.21 consensus estimate. The company achieved a 49% adjusted gross margin. Chairman and CEO Jack Abuhoff said the largest customer represented 37% of quarterly revenues, down from 56% in the first quarter, whilst a Big Tech customer rose to 34% from 17%. The company added a new frontier-lab customer. President Rahul Singhal highlighted new programmes spanning agentic AI, model evaluation, cybersecurity, and physical AI. Innodata is developing motion-capture capabilities and released two public benchmarks. The company reiterated full-year 2026 revenue growth guidance of 40% or more.
Innodata shares fell 5% on Friday following the announcement of a $300 million at-the-market equity programme, sparking dilution concerns amongst investors. The decline overshadowed strong second-quarter results that beat analyst expectations. The AI data and digital services company reported Q2 revenue of $92.1 million, up 58% year-on-year. Diluted earnings per share rose to $0.41 from $0.20, whilst adjusted EBITDA climbed 92% to $25.4 million. The equity programme could result in approximately 4.15 million new shares at recent prices, representing a potential 12% increase to the existing share count of 34.4 million. Management reaffirmed guidance for more than 40% revenue growth in 2026, implying annual revenue of roughly $352 million.
Innodata reported record second quarter 2026 results with revenue of $92.1 million, up 58% year-over-year and beating consensus by 7%. Adjusted EBITDA reached $25.4 million, exceeding consensus by 50%. The company posted net income of $14.4 million, or $0.41 per diluted share, compared to $7.2 million in the prior year period. Adjusted gross margin expanded to 49%, nine points above the company's 40% target. Cash and short-term investments totalled $250.4 million at quarter end. Innodata announced a leadership transition effective 30 September 2026. Rahul Singhal will become president and chief executive officer, whilst founder Jack Abuhoff will transition to executive chairman. The company reiterated full-year 2026 revenue growth guidance of 40% or more year-over-year.