Software Engineer
Develop, deploy, and maintain Python-based applications powered by LLMs and other Gen AI models (e.g., GPT4, Claude, Gemini, Mistral). Implement RAG (Retrieval-Augmented Generation) pipelines using vector databases (e.g., Pinecone, FAISS, Weaviate). Build prompt templates, manage system prompts, and experiment with prompt engineering techniques for performance tuning. Integrate LLMs with APIs, microservices, databases, and user-facing apps. Optimize latency, throughput, and cost-efficiency for inference workloads, including caching, batching, or quantization where applicable. Collaborate with ML and DevOps teams to enable CI/CD, model versioning, and monitoring for AI applications. Work with evaluation frameworks (e.g., LangChain, Weights & Biases, Trulens) to measure output quality (e.g., relevance, coherence, safety). Troubleshoot and enhance LLM behavior, including addressing hallucinations, bias, and task completion accuracy.
Required Qualifications:
- Bachelors or Masters degree in Computer Science, AI/ML, Engineering, or a related field.
- 3+ years of experience in Python software development, preferably with AI/ML projects.
- Strong knowledge of LLMs, transformers, and modern NLP tools.
- Experience working with LLM APIs (OpenAI, Anthropic, Google Gemini, Cohere, etc.).
- Familiarity with vector databases (e.g., FAISS, Pinecone, Chroma) and semantic search principles.
- Hands-on experience with prompt engineering and LLM orchestration tools (LangChain, LlamaIndex, Haystack, etc.).
- Solid understanding of REST APIs, containers (Docker), and cloud platforms (AWS/GCP/Azure).
Preferred Qualifications:
- Experience with chatbot frameworks, agent-based LLM systems, or multi-agent orchestration.
- Understanding of tokenization, context window management, and cost optimization strategies for LLMs.
- Exposure to multimodal AI (e.g., text-to-image, speech-to-text, OCR) and computer vision libraries.
- Familiarity with evaluation techniques for AI output (BLEU, ROUGE, semantic similarity, etc.).
- Contributions to open-source LLM or Gen AI frameworks are a strong plus.