Python Developer - LLM/AI Agents

We are a 800+ strong team whose ideas underpin successful exits, IPOs, strong financial results and competitive advantages as well as novel processes, new business models and great user experiences. We focus on e-commerce, mobility, and transport & logistics, while always keeping an eye on emerging segments as well. We rethink processes, business models and architecture, not to mention tech stacks. This approach opens doors for our people to find their own ways and solutions and bring their own ideas to life, learning and advancing both within and across teams and departments. 
We’re hiring a Python Developer to help us build intelligent, autonomous AI agents using Large Language Models (LLMs). You'll design and orchestrate complex agent behaviors using LangChain, while operating models across cloud, local, and edge environments. This role bridges cutting-edge LLM research with real-world deployment — from prompt engineering and fine-tuning to monitoring and observability. You'll work at the frontier of LLMs, agent architecture, and real-world deployment.

🛠️ What You’ll Do

• Architect and implement modular LLM-driven agents that handle reasoning, planning, and tool usage.
• Operate and integrate LLMs across environments — from cloud-hosted models (e.g., OpenAI, Anthropic, Azure, Bedrock) to local/edge deployments (e.g., llama.cpp, vLLM, Ollama).
• Fine-tune and customize open-source LLMs using LoRA, QLoRA, PEFT, and other efficient tuning strategies.
• Monitor, evaluate, and debug agent performance using LangSmith, OpenInference, Promptfoo, or custom observability tools.
• Build retrieval-augmented generation (RAG) pipelines using vector stores like FAISS, Chroma or Pinecone.
• Orchestrate queries
• Collaborate with ML engineers, DevOps, and product teams to move prototypes into stable, production-ready systems.

✅ Must-Have Skills

• Strong experience in Python 3.x, with a focus on modular, well-tested code.
• Hands-on experience with LLMs in the cloud (OpenAI, Azure OpenAI, Claude, Bedrock, Hugging Face Inference Endpoints, etc.).
• Experience operating LLMs locally/on-device (e.g., using Ollama, llama.cpp, vLLM, GGUF/GPTQ models).
• Solid understanding of LLM agent frameworks (LangChain, CrewAI, AutoGPT) and prompt chaining.
• Practical experience fine-tuning models (e.g., using Hugging Face Transformers, PEFT, LoRA/QLoRA).
• Proficiency in using observability and evaluation tools (e.g., LangSmith, Promptfoo, TruLens) to trace, log, and improve performance.
• Experience with embeddings, vector databases, and contextual memory systems.

🧩 Nice to Have

• Experience deploying LLMs to Kubernetes, serverless platforms, or edge devices (e.g., Jetson, mobile, Raspberry Pi).
• Familiarity with quantization and optimization for low-latency inference.
• Built or contributed to multi-agent systems or long-context memory chains.
• Experience with model monitoring, alerting, and auto-evaluation pipelines in production.

Why you’ll love working here

• Innovative Environment: Be part of a forward-thinking team that values creativity and continuous learning.
• Growth Opportunities: Enhance your career with opportunities for professional development and leadership.
• Competitive Benefits: Enjoy a competitive salary package and benefits tailored to support your well-being.
If you are a passionate technical leader ready to make a significant impact in a client-centric environment, we’d love to hear from you. Apply now and join us on our exciting journey!
Benefits include:
• 🏥 Health Insurance: Comprehensive group health insurance.
• 📊 Social Security: Contributions to the Social Security Office (SSO).
• 💰 Bonus: Attractive 13th-month bonus based on company's performance.
• 🌈 Work Environment: Vibrant and innovative work environment.
• 📈 Career Growth: Opportunities for career growth and development.
• 🚊 Convenient Location: Right at BTS Phrakhanong.