Applied AI Engineer

Keka Technologies Private Limited; Mohali; On-site; INR 4,000,000 - 7,500,000; Full time

Job summary

Keka Technologies Private Limited in Mohali, India, seeks an ML/AI Engineer with 3–6 years of experience to design and deploy AI systems in production. You will bridge research and engineering, building scalable, reliable LLM-driven solutions with strong Python skills and a focus on cost efficiency.

You will implement RAG pipelines, leverage vector databases, and collaborate across backend and frontend teams to deliver robust AI services with good observability and governance.

Qualifications

• Strong Python programming skills and production experience with AI systems.
• Hands-on experience with PyTorch and/or TensorFlow in production environments.
• Practical experience with Hugging Face Transformers and modern LLM APIs.
• Solid understanding of tokenization, embeddings, semantic similarity, and vector search.
• Experience building and optimizing Retrieval-Augmented Generation (RAG) pipelines.
• Familiarity with containerization and deployment in Kubernetes environments.
• Experience with MLOps tooling for model versioning, deployment, and monitoring.

Responsibilities

• Design and deploy RAG pipelines using embedding models and vector databases.
• Develop LLM-powered applications with orchestration frameworks.
• Implement agent-based systems for multi-step reasoning and tool invocation.
• Fine-tune models using LoRA and PEFT techniques.
• Build structured output pipelines with schema validation and function-calling APIs.
• Create automated evaluation pipelines with benchmarks and test prompts.
• Monitor latency, token usage, throughput, and cost across AI workflows.
• Add observability with logging, tracing, and prompt telemetry.
• Collaborate with backend/frontend teams to integrate AI services via REST or async workflows.
• Ensure robustness against hallucinations and prompt injections.

Skills

Python

PyTorch

TensorFlow

LLM APIs

Embeddings

RAG pipelines

Docker

Kubernetes

API design

Microservices

Tools

FAISS

Pinecone

Weaviate

Chroma

LangChain

LlamaIndex

Job description

Position Overview

We are hiring a Machine Learning / AI Engineer with 3–6 years of experience designing and deploying AI systems in production environments. This role requires strong Python expertise, hands‑on experience with generative AI architectures, and the ability to productionize LLM driven systems with reliability, observability, and cost efficiency in mind. The ideal candidate can bridge research and engineering, translating AI concepts into scalable systems.

Key Responsibilities

• Design and deploy Retrieval-Augmented Generation (RAG) pipelines using embedding models and vector databases such as FAISS, Pinecone, Weaviate, or Chroma.
• Develop LLM-powered applications using frameworks such as LangChain, LlamaIndex, AutoGen, or similar orchestration frameworks.
• Implement agent-based systems that support multi-step reasoning, tool invocation, and structured workflows.
• Fine-tune or adapt models using techniques such as LoRA (Low-Rank Adaptation) and parameter-efficient fine-tuning (PEFT).
• Build structured output pipelines using schema validation and function-calling APIs.
• Develop automated evaluation pipelines, including benchmarking datasets, regression testing for prompts, and output quality scoring.
• Monitor and optimize latency, token usage, throughput, and cost across AI workflows.
• Build observability layers for AI systems including logging, tracing, prompt telemetry, and output validation.
• Collaborate with backend and frontend teams to integrate AI services into production systems via REST APIs or asynchronous workflows.
• Ensure robustness against hallucinations, prompt injection, and adversarial inputs through validation logic and contextual constraints.

Required Technical Skills

• Advanced proficiency in Python.
• Hands‑on experience with PyTorch and/or TensorFlow.
• Practical experience with Hugging Face Transformers and modern LLM APIs.
• Strong understanding of tokenization, embeddings, semantic similarity, and vector search.
• Experience building and optimizing RAG pipelines.
• Familiarity with containerization (Docker) and deployment in Kubernetes environments.
• Experience with MLOps tooling for model versioning, deployment, and monitoring.Strong API design and microservices fundamentals.

Preferred Qualifications

• Experience deploying open-source models in production environments.
• Experience with distributed training or inference optimization.
• Familiarity with cloud ML platforms (AWS SageMaker, GCP Vertex AI, Azure ML).
• Knowledge of evaluation frameworks and human-in-the-loop validation systems.

Ideal Candidate Profile

• Execution-oriented with a strong focus on production readiness.
• Able to reason deeply about model behavior, trade‑offs, and system constraints.
• Comfortable iterating quickly while maintaining engineering rigor.
• Collaborative and able to clearly explain complex AI systems to cross‑functional teams.
• Driven by measurable product impact rather than experimentation alone.