Ai/ml Intern

Join us as an AI/ML developer to help train our AI chatbot with advanced emotional recognition capabilities

**Responsibilities**:

- Train and fine-tune the LLaMA 405B model to improve chatbot responses, focusing on emotional recognition and cultural sensitivity.
- Collaborate with mental health experts to adapt CBT content into AI-driven conversational flows.
- Develop and implement NLP algorithms to create personalized and context-aware interactions for users.
- Integrate AI systems with healthcare platforms while ensuring data privacy and compliance with regulations.
- Conduct model evaluations, run A/B tests, and make iterative improvements based on user interactions and feedback.
- Work closely with the AR/VR team to incorporate AI functionalities into the VR experience.
- Build and optimize datasets to enhance the training of the LLaMA model for diverse user needs, ensuring inclusivity and cultural relevance.
- Develop methods for sentiment analysis and real-time emotional tracking, enabling the chatbot to respond empathetically and adaptively to user states.
- Collaborate with data scientists and clinical psychologists to ensure model predictions align with therapeutic best practices.
- Lead technical workshops and provide mentorship to junior AI developers.

**Requirements**:

- Degree or Diploma in Computer Science, Data Science, AI/ML, or a related field.
- Strong Python skills, experience with LLMs (preferably LLaMA or similar), expertise in NLP, understanding of integrating AI with healthcare settings, and knowledge of privacy regulations (e.g., GDPR).

**Internship: INR 5,000 to 15,000 stipend,** only for the current Students and Recent graduates

**Salary Range: INR 2,00,000 to INR 18,00,000**per annum depending on experience and expertise.

**Assignment for AI/ML Developer Role**

**Assignment Overview**:
**Project: Emotional Sentiment Analysis and Adaptive Response System**

Objective: Develop a prototype chatbot capable of identifying the emotional state of a user based on a conversation and responding with culturally relevant and empathetic responses. This exercise will help assess your ability to work with NLP models and adapt their output to user needs in a culturally sensitive context.

**Tasks**:

- Data Collection and Preprocessing
- Find and prepare a dataset containing conversational data that includes emotional indicators. Annotate this data for emotional states such as sadness, anxiety, stress, and joy.
- Clean the data to handle missing values, standardize text (e.g., removing stop words, punctuation), and prepare it for training purposes.
- Model Training
- Train a language model using an existing pre-trained model (preferably a version of GPT-3 or LLaMA) to perform sentiment analysis on the dataset prepared.
- Fine-tune the model to classify emotional states with an emphasis on cultural context and linguistic nuances.
- Response Generation
- Create a response generation system that produces empathetic responses based on the identified emotional state. The responses should be tailored to cultural and contextual cues, reflecting appropriate support for each emotional state.
- Integration and Testing
- Integrate the sentiment analysis module with the response generation module to create a cohesive chatbot prototype.
- Test the chatbot prototype with a set of simulated conversations and document the results.
- Analyse the chatbot’s performance, including how accurately it identifies emotional states and how well its responses match the context and cultural expectations.
- Documentation
- Prepare documentation summarizing your approach, data preparation techniques, model choices, challenges faced, and results.
- Include a reflection section on how you would further improve the model’s ability to understand cultural sensitivities.

**Submission Requirements**:

- Provide the code, dataset (or dataset link), and a report that includes the results and discussion on how you approached the problem.
- The submission should be in a GitHub repository, which will allow us to view both the code and the documentation conveniently.

**Evaluation Criteria**:

- Technical Skills: How well the model performs sentiment analysis, and the quality of the chatbot’s responses.
- Problem-Solving: Ability to tackle data preprocessing challenges and adapt a pre-trained model to the task.
- Creativity and Cultural Sensitivity: Quality of generated responses, with emphasis on cultural adaptation and empathy.
- Code Quality and Documentation: Clear, maintainable code, along with thorough and understandable documentation.

**Job Types**: Full-time, Part-time, Permanent, Fresher, Internship, Contractual / Temporary, Volunteer

Pay: ₹200,000.00 - ₹1,800,000.00 per year

**Benefits**:

- Flexible schedule
- Work from home

Schedule:

- Day shift
- Monday to Friday
- Morning shift
- Night shift
- UK shift
- US shift
- Weekend availability

Application Question(s):

- Ready to complete an initial assignment as a a