About the Role

As an AI/NLP Engineer on the vFairs Tech Team, you will design, build, and maintain the natural‑language driven form‑generation and data‑retrieval engine that powers our next‑generation “Lovable” and “Replit‑style” experiences. You’ll collaborate closely with full‑stack developers, product managers, and DevOps to deliver a seamless pipeline from user prompt → responsive HTML/CSS → chained form flows → database persistence → generic query APIs.

Key Responsibilities

• 
  Prompt Engineering & Generation

• 
  Craft and iterate high‑precision prompts that translate NL requirements into valid, responsive HTML/CSS form definitions.

• 
  Implement automated post‑processing and code‑sanitization layers to ensure production readiness.

• 
  Retrieval‑Augmented Generation (RAG)

• 
  Design vector‑search pipelines using FAISS/Pinecone/Weaviate to index form metadata, user‑generated docs, and knowledge bases.

• 
  Integrate retrieval calls into LLM prompts to ground outputs in up‑to‑date organizational data.

• 
  Agent & Orchestration Frameworks

• 
  Build multi‑step “agentic” flows leveraging frameworks such as LangChain, LlamaIndex, or Microsoft Semantic Kernel.

• 
  Enable dynamic tool‑calls, decision logic, and error‑handling within generation pipelines.

• 
  Model Fine‑Tuning & Training

• 
  Fine‑tune open‑source LLMs (e.g. Vicuna, Alpaca) and implement instruction‑tuning workflows using Hugging Face Accelerate.

• 
  Curate datasets of prompt–completion pairs and set up basic RLHF or reward‑model loops for iterative improvement.

• 
  Embeddings & Similarity Search

• 
  Select and evaluate embedding models (OpenAI “text-embedding-3”, Sentence‑Transformers) for downstream tasks.

• 
  Implement nearest‑neighbor search with dimensionality reduction and performance optimization.

• 
  Evaluation & Monitoring

• 
  Define metrics (accuracy, diversity, safety) for generated code quality and monitor drift over time.

• 
  Set up logging, A/B experiments, and dashboards to track system performance and user satisfaction.

• 
  API & System Integration
• 
  Collaborate with back‑end teams to expose generation and retrieval endpoints (REST or GraphQL).
• 
  ​
  Ensure secure handling of prompts and responses, manage rate limits, and implement retry logic.

Required Qualifications

• 
  BS/MS in Computer Science, AI/NLP, Computational Linguistics, or related field.
• 
  2+ years of hands-on experience with large‑language models and NLP pipelines.
• 
  Proficient in Python and familiar with frameworks: Hugging Face Transformers, LangChain or similar.
• 
  Deep understanding of RAG architectures and vector databases (FAISS, Pinecone, etc.).
• 
  Experience fine‑tuning LLMs, instruction‑tuning, and basic RLHF workflows.
• 
  Strong skills in prompt engineering and code sanitization.
• 
  Ability to translate product requirements into technical designs and prompts.
• 
  Strong collaboration skills to work across cross‑functional teams.