Scribd
Upload
4 views

Assignment for AI

The assignment for the AI/NLP Engineer role involves building a prototype system for generating and refining educational content, focusing on user prompts, content quality, and bias mitigation. Deliverables include a content generation pipeline, refinement module, bias detection mechanisms, and comprehensive technical documentation. Candidates will be evaluated on technical proficiency, problem-solving skills, scalability, content quality, deployment, and communication, with a phased submission process culminating in a live demonstration.

Uploaded by

devanshu9089
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
4 views

Assignment for AI

The assignment for the AI/NLP Engineer role involves building a prototype system for generating and refining educational content, focusing on user prompts, content quality, and bias mitigation. Deliverables include a content generation pipeline, refinement module, bias detection mechanisms, and comprehensive technical documentation. Candidates will be evaluated on technical proficiency, problem-solving skills, scalability, content quality, deployment, and communication, with a phased submission process culminating in a live demonstration.

Uploaded by

devanshu9089
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Assignment for AI/NLP Engineer Role

Objective:
Evaluate the candidate’s technical expertise in designing, implementing, and integrating AI and NLP
components for a sophisticated content generation and refinement platform.

Task Description:

You are tasked with building a prototype system for generating and refining educational content
tailored to a specific subject or grade level. The system should demonstrate your ability to:

1. Process user prompts to generate coherent and contextually relevant content.

2. Refine the generated content to meet high standards of readability, coherence, and
inclusivity.

3. Incorporate mechanisms to ensure fairness and mitigate biases in the generated content.

4. Develop modular and scalable pipelines for AI/NLP workflows.

Deliverables:

1. Prompt-to-Content Generation:

o Build a pipeline that accepts user prompts and generates educational content using
a pre-trained transformer-based language model (e.g., GPT-3.5, GPT-4, or similar).

o Ensure that the content aligns with general curriculum standards.

2. Content Refinement Module:

o Implement a module to refine the generated content for linguistic clarity, coherence,
and contextual alignment.

o Apply advanced techniques like reinforcement learning for content optimization


(optional but preferred).

3. Bias Detection and Mitigation:

o Include a mechanism to detect and mitigate biases in the generated content,


ensuring inclusivity and fairness.

4. Pipeline Design and Modularity:

o Structure the system to include modular pipelines for content generation,


refinement, and evaluation.

o Document how the pipeline can be extended to support additional tasks or modules
in the future.
5. Technical Documentation:

o Provide a detailed report explaining your architecture, techniques, and tools used.

o Include metrics for evaluating the content quality and system performance.

6. Deployment:

o Deploy the prototype locally or on a cloud platform (e.g., AWS, GCP).

o Include instructions for running the system.

Evaluation Criteria:

1. Technical Proficiency:

o Expertise in using AI frameworks and libraries such as Hugging Face Transformers,


TensorFlow, or PyTorch.

o Proficiency in Python and NLP-specific libraries like spaCy or NLTK.

2. Problem-Solving Skills:

o Innovative approaches to content generation, refinement, and bias mitigation.

3. Scalability and Modularity:

o Demonstrated ability to build a scalable and modular system.

4. Content Quality:

o Metrics like coherence, relevance, inclusivity, and readability of generated content.

5. Deployment and Usability:

o Effective deployment and clarity of user instructions.

6. Communication:

o Clear, comprehensive documentation of the work.

Assignment Workflow:

1. Phase 1: Proposal Submission

o Outline your planned approach, tools, and techniques in a brief proposal.

o Submission deadline: 1/02/2025.

2. Phase 2: System Implementation and Report

o Submit the implemented prototype along with the technical report.

o Submission deadline: 1/02/0225.


3. Phase 3: Live Demonstration and Q&A

o Present your system and demonstrate its capabilities.

o Discuss the rationale behind your design decisions.

Notes:

• The use of open-source models and libraries is encouraged, but the candidate must justify
their choices.

• Collaboration is not allowed; this is an individual assignment.

• The solution should focus on modularity and scalability, preparing for future
enhancements.

You might also like