AI Developer (Agentic AI Frameworks, Computer Vision & LLMs)
Dear Candidate.
Greetings from 3 Kubed Consulting India Pvt Ltd.!!
We are seeking a highly skilled and experienced AI Developer (Agentic AI Frameworks, Computer Vision & LLMs)” for one of our prestigious clients for Bangalore location. The brief JD is as below.
• EXP: 3+ years
• REL EXP: 3+ YRS
• Location: Bangalore
• NP: Ideally Immediate, max 30 days
• Interview: 2/3 virtual rounds, one python coding round
• Work mode: REMOTE
• Employment Mode: PMT with CAI
Our client is seeking an AI Developer who specializes in agentic AI frameworks—LangChain, LangGraph, CrewAI, or equivalents—and who can take both vision and language models from prototype to production. You will lead the design of multi agent systems that coordinate perception (image classification & extraction), reasoning, and action, while owning the end to end deep learning life cycle (training, scaling, deployment, and monitoring).
Key Responsibilities
Agentic AI Frameworks (Primary Focus):
• Architect and implement multi agent workflows using LangChain, LangGraph, CrewAI, or similar.
• Design role hierarchies, state graphs, and tool integrations that enable autonomous data processing, decision making, and orchestration.
• Benchmark and optimize agent performance (cost, latency, reliability).
Image Classification & Extraction:
• Build and fine tune CNN/ViT models for classification, detection, OCR, and structured data extraction.
• Create scalable data ingestion, labeling, and augmentation pipelines.
LLM Fine Tuning & Retrieval Augmented Generation (RAG):
• Fine tune open weight LLMs with LoRA/QLoRA, PEFT; perform SFT, DPO, or RLHF as needed.
• Implement RAG pipelines using vector databases (FAISS, Weaviate, pgvector) and domain specific adapters.
Deep Learning at Scale:
• Develop reproducible training workflows in PyTorch/TensorFlow with experiment tracking (MLflow, W&B).
• Serve models via TorchServe/Triton/KServe on Kubernetes, SageMaker, or GCP Vertex AI.
MLOps & Production Excellence:
• Build robust APIs/micro services (FastAPI, gRPC).
• Establish CI/CD, monitoring (Prometheus, Grafana), and automated retraining triggers.
• Optimize inference on CPU/GPU/Edge with ONNX/TensorRT, quantization, and pruning.
Collaboration & Mentorship:
• Translate product requirements into scalable AI services.
• Mentor junior engineers, conduct code and experiment reviews, and evangelize best practices.
Minimum Qualifications:
• ./. in Computer Science, Electrical Engineering, Applied Math, or related discipline.
• 5+ years building production ML/DL systems with strong Python & Git.
• Demonstrable expertise in at least one agentic AI framework (LangChain, LangGraph, CrewAI, or comparable).
• Proven delivery of computer vision models for image classification/extraction.
• Hands on experience fine tuning LLMs and deploying RAG solutions.
• Solid understanding of containerization (Docker) and cloud AI stacks (AWS/Azure).
• Knowledge of distributed training, GPU acceleration, and performance optimization.
With regards,
For 3 Kubed Consulting India Pvt Ltd