Data Scientist

Company Overview:Galaxy Office Automation Pvt. Ltd. is a trusted enterprise technology partner with 30+ years of experience in delivering secure, large-scale IT systems across India’s top enterprises and government institutions. As we scale into the next frontier of AI-driven intelligence, we’re building a new stack of AI-native services, products, agents, swarms, and conversational assistants that bring powerful, modular intelligence to real-world enterprise workflows – while evolving towards an AI-Factory model to enable scalable, reusable, and composable intelligence across use cases.Our goal is not buzzwords—it’s capability. We focus on deploying reliable, composable, and production-grade AI systems that can plug directly into existing ecosystems and deliver immediate business value.Role Summary  : We’re looking for a Data Scientist / AI Engineer / Machine Learning Engineer who can own systems end to end and contribute from the ground up . In this role, you’ll architect and deploy real-world AI solutions using LLMs, predictive modeling, multimodal intelligence , and FastAPI-based microservices . You’ll work across core platform modules and client-specific projects—bridging the gap between cutting-edge AI and enterprise-grade deployment.What You’ll Work On● Multi-Agent Collaboration, Reasoning, Memory & Human Alignment Build intelligent agents and swarms with multi-agent collaboration, reasoning, planning, memory, and alignment with human feedback. Use protocols such as MCP and A2A and frameworks such as LangChain and Crew AI .● Retrieval-Augmented Generation (RAG) Develop hybrid pipelines using vector databases (FAISS, Qdrant, Pinecone) and transformer-based generation models● Multimodal AI (Language + Vision + Audio + Video) Build systems that process and combine intelligence across language, vision, audio, and video.● Predictive Modeling Build predictive pipelines using deep learning architectures (e.g., LSTMs, CNNs, RNNs), transformer-based models (e.g. openAI, LLama, Qwen, Mistral) , and ensemble methods (e.g., XGBoost, LightGBM, Random Forests). Emphasize modeling depth, generalization, interpretability, maintainability and dynamic improvements.● Conversational Assistants Develop conversational assistants with advanced capabilities such as model based recommendations, user-query based what-if scenario analyses and continuous improvements based on memory and human feedback.● FastAPI-Based Backend APIs Wrap agents and models into versioned, secure, and production-grade FastAPI microservices .● Model Lifecycle Management Track, evaluate, and manage model lifecycles using MLflow, DVC, and internal governance tools.● Data Engineering & Integration Ingest and transform data from SQL and NoSQL (e.g. MongoDB) sources, APIs, and distributed pipelines.Required Skills and Qualifications● 1–4 years of experience in AI/ML product development or applied data science● Strong Python skills: Pandas, NumPy, scikit-learn, Transformers, PyTorch/TensorFlow● Hands-on experience with LLMs (OpenAI, Mistral, Claude, Llama, Deepseek, Gemini, etc.), LangChain, prompt engineering, and integration with real-world use cases● Proven experience building agentic systems, including reasoning agents and multi-agent collaboration● Deep expertise in predictive modeling using transformers, deep learning and ensemble methods● Familiarity with image, audio, and video model development● Familiarity with model monitoring and re-training● Exposure to both SQL and NoSQL databases● Experience building and deploying Python based backend APIs using FastAPI● Proficiency with Git workflows, CI/CD, and modular code development● Strong communication, documentation, and architectural thinkingDesirable Skills● Candidates with experience in reinforcement learning (including RLHF) will be preferred.● Prior exposure to supervised fine-tuning (SFT) and parameter-efficient tuning approaches such as LoRA and QLoRA is desirable.● Familiarity with workflow orchestration tools like Airflow, Celery, Prefect, or Dagster will be advantageous.● Experience building autoscaling and serverless architectures using AWS Lambda, ECS, or EKS is a plus.● Candidates with an expertise in backend engineering with a focus on microservice optimization will be preferred.● Candidates with experience in containerization and orchestration using Docker and Kubernetes will be valued.● Hands-on knowledge of designing conversational UI or chatbot pipelines is beneficial.● Prior experience working with big data tools like Spark, Hive, or Hadoop is desirable.● Experience deploying AI systems in BFSI, healthcare, e-commerce, or government environments will be an added advantage.What You’ll Gain● End-to-End Ownership from design to deployment ● Agent-First Ecosystem Exposure ● Product + Custom Work across client and platform needs ● Leadership Track into applied AI architecture ● Live Enterprise Impact across sectorsIf you think in embeddings, talk in APIs, and believe human-aligned, reasoning-first multi-agent systems are the future—join Galaxy Office Automation and help build them.