Senior Data Scientist Engineer

We are seeking an experienced Agentic Automation Engineer to join our team. The ideal candidate will have expertise in Conversational AI, specifically working with LLM-based conversational AI models. This role requires strong configuration skills and the ability to problem-solve while being aware of the constraints of such models.Essential functions

Key Responsibilities

Develop and implement LLM-based conversational AI solutions

Configure and optimize AI models for specific use cases

Troubleshoot and resolve issues related to AI model performance

Stay up to date with the latest advancements in LLM technologies and apply them to our projects

Qualifications

Experience with the following LLM-related technologies: ( At least 5 LLM technologies )

1. Chunking
2. Document Parsing and OCR
3. Document Parsing with VLMs (Vision Language Models)
4. Function Calling with LLMs
5. Building Agentic Chains
6. Memory Handling (Conversational Memory) and Tracking / Tracing Software Agents
7. Map Reduce applied to RAG
8. Retrieval Augmented Generation
9. Traditional Search (BM25, NER based parsers, Keyword based search index)
10. Semantic Search (Embeddings, Embedding models)
11. Quantized Embeddings
12. Reference Management
13. Re-ranking references
14. Re-ranking retrieved chunks
15. Multimodal RAG
16. Cost reduction techniques for LLMs
17. Speculative Decoding
18. Speech to Text using Whisper or similar models
19. Text to SQL/Text to Pandas
20. Monte Carlo Tree Search, Tree of Thoughts, Chain of Thoughts, Reasoning Enhancement Techniques for LLMs

Machine Learning Skills

Experience with the following ML techniques: ( At least 3 ML Skills )

1. Fine Tuning using LoRA
2. Merging multiple LoRA adapters using MergeKit
3. Quantising LLMs
4. Quantising LoRA adapters
5. Creating Table Schemas suitable for LLMs
6. Code Folding for Coding LLMs
7. Implementation of Code Interpreters (E.g., E2B Dev)

Required Knowledge

Understanding of differences between code-focused LLMs, Completion LLMs, and Chat variants of LLMs

Framework Experience ( At least 3 frameworks )

1. E2B Dev - Code Interpreters
2. DSPY - Automation of Prompt Engineering
3. Lang chain
4. OpenAI APIs, Claude APIs
5. Guidance (For Prompt Following reliability)
6. OpenAI whisper for STT