Agentic AI Technical Architect

Job description AI Research Engineer

Contract: 3-6 Months+

Remote

Role Description

We are seeking an AI Research Engineer to improve the performance and reliability of our conversational AI assistant.

The role focuses on advancing LLM capabilities including reasoning, retrieval-augmented generation (RAG), fine-tuning, and structured outputs.

You will research, experiment, and implement improvements that directl...y impact production AI systems.

Responsiblities

• Research and evaluate techniques to improve LLM reasoning, accuracy, and efficiency

• Design and run experiments with baselines, ablations, and statistical analysis

• Build evaluation frameworks and benchmarks for model performance across key tasks

• Improve RAG pipelines including embedding models, chunking strategies, re-ranking, and context optimization

• Explore fine-tuning methods such as LoRA, RLHF, and DPO for domain-specific models

• Identify and reduce failure modes such as hallucination, instruction drift, and reasoning errors

• Prototype advanced techniques like chain-of-thought reasoning, tool use, and multi-agent coordination

• Translate research insights into production-ready improvements and technical documentation

Qualification

• 4–5 years of experience in AI/ML research or applied AI engineering

• Strong understanding of LLM architectures, training dynamics, and inference optimization

• Hands-on experience with fine-tuning methods such as LoRA, RLHF, or DPO

• Proficiency in Python and ML frameworks such as PyTorch and HuggingFace

Must Have

• Experience building and evaluating LLM-based systems

• Strong knowledge of RAG pipelines, embeddings, and retrieval systems

• Experience designing reproducible experiments and evaluation benchmarks

• Ability to analyze model performance and implement improvements

Nice to have

• Published research in NLP, LLMs, or information retrieval

• Experience with Gemini, Claude, GPT, or similar model APIs

• Experience building automated evaluation and benchmarking systems

• Familiarity with model optimization techniques such as distillation or quantization

• Experience with agentic AI frameworks such as LangGraph or similar orchestration systems

Job Type: Full-time

Work Location: Remote Show full description Report this listing Loading...