Grid Dynamics - Generative AI Expert - Agentic AI

We are Hiring :

About the Role :

In this role, you will design, build, fine-tune, and deploy production-grade LLM agents and multi-modal AI applications that solve real-world business challenges.

You will play a key role in shaping agent design, orchestration, and observability, ensuring enterprise-grade scalability, robustness, and performance.

Key Responsibilities :

Model & Agent Design :

- Build multi-agent and multi-modal AI applications/frameworks, ensuring interactivity, latency optimization, failover, and usability.

- Apply advanced design principles for structured outputs, tool usage, speculative decoding, AST-Code RAG, streaming, and async/sync processing.

Hands-on Coding & Development :

- Write, test, and maintain clean, scalable, and efficient Python code for LLMs and AI agents.

- Implement fine-tuning, embeddings, and prompt engineering with a focus on cost, latency,

and accuracy.

- Integrate models with vector databases (Milvus, Qdrant, ChromaDB, CosmosDB, MongoDB).

Performance & Monitoring :

- Monitor and optimize LLM agents for latency, scalability, robustness, and explainability.

- Implement observability and guardrails strategies for enterprise-safe AI deployments.

- Handle model drift, token consumption optimization, and error recovery mechanisms.

Research & Innovation :

- Read, interpret, and implement AI/Agent research papers into practical production-ready solutions.

- Stay ahead of academic and industry trends in Agentic AI, multimodal AI, orchestration

frameworks, and evaluation methodologies.

- Experiment with new AI orchestration tools, evaluation frameworks, and observability platforms (Arize or similar).

Debugging & Issue Resolution :

- Diagnose and resolve model inaccuracies, system integration issues, and performance bottlenecks.

- Apply advanced debugging techniques to troubleshoot deployment errors, data inconsistencies, and unexpected agent behaviors.

Continuous Learning & Adaptability :

- Quickly unlearn outdated practices and adapt to emerging GenAI and Agentic AI technologies.

Required Skills & Experience :

- 5 - 9 years total experience, with 4+ years in NLP, CV, and LLMs.

- Strong expertise in GenAI, LLMs, RAG pipelines, embeddings, and vector databases.

- Proficiency in Python with strong debugging and system design skills.

- Hands-on experience with Agentic AI frameworks (LangChain Agents, AutoGen, CrewAI,

Temporal, DSPy).

- Proven record of production-grade AI deployments (not just POCs).

- Cloud experience : Azure (preferred), AWS, or GCP.

- Knowledge of AI orchestration, evaluation, guardrails, and observability tools (Arize, Weights & Biases, etc.