AI Engineer – LLM Systems & Applied AI

Apply Here: https://forms.gle/Qc6KCRVXVNFYSh86A

Company: Infiheal HealthTech Private Limited

Location: Remote / Hybrid (Mumbai preferred)

Employment Type: Full-Time

Experience: More than 2 years of professional experience (excluding internships)

Compensation: Competitive, based on experience

About Infiheal:

Infiheal is an IIT Bombay-incubated AI mental healthcare startup, recognized in Mann Ki Baat by the Hon'ble Prime Minister of India.

With 1M+ users across 165+ countries, our flagship products, Healo and DuoChat, leverage multi-LLM architectures, advanced RAG, psychological intelligence, and AI safety systems to deliver personalized mental health support across 93+ languages.

We are looking for a hands-on AI Engineer experienced in building, optimizing, and scaling production-grade AI systems, with an interest in emerging LLM technologies and infrastructure.

Key Responsibilities:

1. LLM Systems & Architecture

- Design and optimize production-grade LLM pipelines, multi-model orchestration, agentic workflows, intelligent model routing, and fallback mechanisms.
- Develop conversational memory, context engineering, structured outputs, tool calling, and state management systems.
- Evaluate proprietary and open-source LLMs for accuracy, latency, reliability, and cost.

2. Advanced RAG & Agentic AI

- Build and optimize RAG pipelines using hybrid search, embeddings, reranking, retrieval evaluation, and context optimization.
- Work with vector databases such as FAISS, Qdrant, Milvus, or Pinecone.
- Develop stateful AI agents with tool integrations, memory, and controlled multi-step execution.

3. AI Safety, Evaluation & Observability

- Develop evaluation pipelines for accuracy, hallucination, bias, safety, and regression testing.
- Implement AI guardrails, risk classification, content moderation, prompt-injection defenses, and escalation mechanisms.
- Monitor model quality, latency, token consumption, errors,



and production reliability.
- Collaborate with psychologists to improve contextual appropriateness, empathy, and safety.

4. Production Engineering & MLOps

- Develop scalable AI microservices and APIs using Python, FastAPI, and asynchronous programming.
- Deploy and manage AI workloads using Docker, Kubernetes/GKE, AWS/GCP, CI/CD, and MLOps/LLMOps practices.
- Optimize real-time conversational AI, streaming responses, and STT/TTS integrations.
- Own AI solutions end-to-end, from architecture and experimentation to deployment and monitoring.

5. AI Infrastructure & Model Optimization

- Understand infrastructure requirements for cloud, self-hosted, and on-premises LLM deployments, including GPU/VRAM sizing, compute resources, and capacity planning.
- Explore model-serving frameworks such as vLLM, SGLang, NVIDIA NIM, or TensorRT-LLM.
- Evaluate inference performance, latency, throughput, GPU utilization, and cost optimization.
- Explore model optimization techniques such as quantization, LoRA/QLoRA, PEFT, fine-tuning, and distillation, based on application requirements.

Required Qualifications:

- More than 2 years of relevant professional experience (excluding internships), with hands-on experience building production AI/ML applications.
- Robust proficiency in Python, FastAPI, and API/microservices development.
- Good understanding of LLMs, transformers, embeddings, tokenization, and inference workflows.
- Hands-on experience with RAG, vector databases, LLM orchestration, and prompt/context engineering.
- Experience with AI evaluation, guardrails, monitoring,



and production troubleshooting.
- Familiarity with PyTorch, Hugging Face Transformers, or equivalent AI/ML frameworks.
- Experience with Docker, cloud infrastructure (AWS/GCP), Git, and CI/CD; working knowledge of Kubernetes.
- Understanding of scalability, performance optimization, reliability, and cost considerations.
- Strong problem-solving skills and ability to independently deliver production-ready AI solutions.

Good to Have:

- Knowledge or experience in self-hosted/on-premises LLM deployment, GPU/VRAM sizing, and infrastructure capacity planning.
- Familiarity with vLLM, SGLang, NVIDIA NIM, TensorRT-LLM, or similar inference frameworks.
- Understanding of quantization, LoRA/QLoRA, fine-tuning, distillation, or model compression.
- Experience with LangGraph, MCP, LlamaIndex, Langfuse, MLflow, Ragas, or equivalent tools.
- Knowledge of CUDA, NVIDIA GPU infrastructure, distributed inference, or inference optimization.
- Experience with voice AI, multilingual NLP, real-time streaming, or speech models.
- Familiarity with HIPAA, GDPR, AI security, or healthcare/mental health applications.
- Open-source contributions or independently developed AI projects.
- Mumbai-based candidates preferred.

Why Join Infiheal?

- Build AI systems supporting 1M+ users across 165+ countries.
- Work on advanced RAG, multi-LLM orchestration, agentic AI, model optimization, and AI safety.
- Gain exposure to self-hosted AI, GPU infrastructure, and emerging LLM technologies.
- Own meaningful technical decisions and collaborate with engineers, psychologists, and researchers.

We're looking for engineers who can go beyond basic LLM API integrations and build reliable, scalable, and intelligent AI systems.

Apply Now:

Application Form: https://forms.gle/Qc6KCRVXVNFYSh86A

Complete the application form with your experience, technical projects, and relevant work.

📌 AI Product Engineer (India)
🏢 Infiheal
📍 India