Location: Hydrabad, India
Type: Full-Time | Immediate Joining Preferred
CTC: Competitive
Company Description
YAL.ai, short for Your Alternative Life, is a cutting-edge platform that transforms communication with AI-driven tools for secure and dynamic interactions. Offering features like instant messaging, group chats, VoIP calls, and multilingual support, YAL.ai provides a seamless and safe digital experience. With a foundation in Zero Trust Architecture, it ensures the highest level of security and privacy through advanced on-device AI. The platform combines innovation and integrity, fostering meaningful connections and creating a vibrant digital space for its users. Located in Hyderabad, YAL.ai is committed to redefining communication with a focus on privacy and security.
Role Overview
We are looking for a Junior NLP Scientist to work on building and improving NLP and LLM-powered features such as smart reply, search and summarization, content moderation, conversational intelligence, and multilingual language processing systems. You will work closely with senior scientists and engineering teams to experiment, prototype, and productionize language models.
Key Responsibilities
- Develop and fine-tune NLP models for classification, tagging, summarization, and conversational tasks
- Work with LLM APIs and open-source models for prompt engineering and task optimization
- Assist in dataset creation, cleaning, augmentation, and annotation workflows
- Implement evaluation pipelines for model accuracy, latency, and robustness
- Support RAG pipelines using vector databases
- Contribute to multilingual NLP initiatives (Indic language support preferred)
- Collaborate with backend engineers to deploy models in production
- Monitor model performance and help improve inference efficiency
Required Qualifications
- Bachelors or Masters in Computer Science, AI, Data Science, Linguistics, or related field
- 1 to 3 years of experience in NLP or ML
- Strong Python programming skills
- Familiarity with PyTorch or TensorFlow
- Hands-on experience with HuggingFace Transformers
- Understanding of tokenization, embeddings, attention, and transformer architectures
- Exposure to knowledge graph concepts (RDF, triples, ontologies)
- Basic knowledge of LLM prompting and evaluation
- Experience with Git and basic MLOps practices
Preferred Skills
- Exposure to RAG systems
- Experience with vector databases such as FAISS or Pinecone
- Familiarity with Indic language datasets
- Experience working with cloud platforms (AWS, GCP, or Azure)
- Knowledge of model optimization and quantization basics
Key Traits
- Strong problem-solving ability
- Curious and research-driven mindset
- Ability to rapidly prototype ideas
- Strong communication and documentation skills
…