Canada
Not specified
March 25, 2026
Job Description
About the Role
Innodata is hiring Prompt Engineers to design and automate data labeling and localization workflows using large language models (LLMs).
You’ll work directly with data scientists, linguists, and product teams to build scalable AI-driven systems that reduce manual annotation effort while maintaining high accuracy and cultural relevance.
What You’ll Do
Advertisement
- Design and implement prompts for:
- Data labeling
- Localization
- Translation workflows
- Prototype and validate LLM-based solutions
- Analyze model performance using KPIs and evaluation metrics
- Optimize prompts through user testing and feedback
- Build and improve AI-driven data pipelines
- Collaborate with cross-functional teams (ML, product, linguistics)
- Develop guidelines and documentation for prompt usage
- Improve automation in annotation and evaluation workflows
Requirements
Core Skills
- Strong understanding of LLMs (transformers, evaluation, bias, reliability)
- Experience using LLMs programmatically
- Python expertise (data processing, NLU, automation)
- Familiarity with:
- APIs (OpenAI, Hugging Face)
- Data formats (JSON, XML, JavaScript)
- Experience with ML frameworks (TensorFlow, PyTorch)
- Understanding of data pipelines and automation systems
Advertisement
Additional Requirements
- Knowledge of localization and cultural nuances
- Strong analytical and problem-solving skills
- Ability to communicate technical insights clearly
- High attention to detail
Preferred
- 2+ years in prompt engineering or AI/ML roles
- Experience with annotation tools (e.g., Labelbox)
- Familiarity with cloud platforms and deployment
- Multilingual knowledge
Education
- Bachelor’s degree or higher in:
- Computer Science
- AI / Machine Learning
- Linguistics or related field
Why This Role Matters
- Work on real-world LLM automation systems
- Reduce human effort in large-scale AI pipelines
- Contribute to high-impact AI infrastructure used globally
Apply Now
You will be redirected to the company's website to complete your application.
Job Summary
Categories
Explore other opportunities in related specializations:
Related Jobs
View All
Innodata
Discover more opportunities that match your skills and interests.
Frequently Asked Questions
Is Innodata legitimate?
Yes, Innodata is a legitimate, publicly traded company (NASDAQ: INOD) that has been in operation since 1988. Listed on RemoWork, Innodata serves major technology companies and enterprises worldwide. As always, we recommend verifying details on their official website and checking user reviews before applying.
How do I get started with Innodata?
Visit Innodata's official careers page to browse open positions. Create an account, submit your application with relevant qualifications and experience, and complete any required assessments (typically a General English test and Versant proficiency test). Some roles may involve additional technical evaluations or interviews.
Does Innodata offer remote work?
Yes, Innodata offers fully remote positions, particularly for data annotation, AI training, and Generative AI Associate roles within the US (some state exclusions may apply). The company also has on-site positions at offices in the Philippines, India, Sri Lanka, and other global locations.
How does Innodata pay its workers?
Innodata offers competitive compensation that varies by role, experience level, and location. Full-time employees receive standard payroll with benefits. Freelance and contract workers are compensated based on project terms. Specific payment details are discussed during the hiring process.
What skills do I need to work with Innodata?
Required skills vary by role. Data annotators need strong English, attention to detail, and the ability to follow complex guidelines. AI engineers need Python proficiency, ML framework experience, and relevant degrees. All roles value analytical thinking, adaptability, and a passion for AI and data quality.
What types of annotation tasks does Innodata offer?
Innodata provides a comprehensive range of annotation services including image annotation, text classification, audio transcription, video labeling, semantic segmentation, bounding box annotation, named entity recognition, sentiment analysis, optical character recognition, entity linking, syntactic parsing, and LLM response evaluation. Tasks cover both structured and unstructured data across 40+ languages.
What is the hiring process like at Innodata?
The hiring process typically includes submitting an application, completing a General English proficiency test and a Versant assessment (evaluating spoken and written English), followed by interviews assessing educational background, professional experience, and domain expertise. Some roles include practical assessments or annotation trials. The process evaluates both qualifications and cultural fit.
Is Innodata a publicly traded company?
Yes, Innodata is publicly traded on the NASDAQ stock exchange under the ticker symbol INOD. Founded in 1988, the company has a long track record in data services and has expanded significantly into AI data engineering. Being publicly traded means greater transparency through regular SEC filings and financial disclosures.