BioSky

Job Description: Founding Team Member - Machine‑Learning Scientist Location: New Delhi Role: Full-Time, Onsite About BioSky: BioSky Space Innovations Pvt. Ltd. is an early-stage spacetech startup focused on geospatial modeling and AI-driven analytics . BioSky harnesses space intelligence to power global industries at scale , delivering business‑critical insights through our space‑tech platform. We integrate multi-source satellite imagery, weather data, and ground observations to build physics-informed geospatial models for forecasting, anomaly detection, and pollutant pathway tracing. Our solutions deliver high-accuracy, location-specific intelligence for sectors like renewable energy, water resources, climate resilience, and defence —helping organizations reduce losses, optimize operations, and make data-driven decisions. We are a funded and IIT Delhi backed startup and selected among top 15 global space-tech startups at Venture Catalyst Space Australia, AWS Space Accelerator Tacc+ and Taiwan International Spacetech 2025. About the role: Join BioSky as a founding Machine‑Learning Engineer focused on Earth‑observation imagery. You will build and deploy computer‑vision and neural‑network models that convert multi‑sensor satellite data into actionable insights. The role offers a stipend and equity, with significant scope to shape both the technology and the company’s growth. We are looking for Machine-Learning Scientists with 2-3 years of hands-on experience who are eager to learn, adapt, and grow rapidly in a high-velocity startup environment. You should be driven by curiosity, thrive on solving open-ended problems, and be ready to take ownership of end-to-end systems. This role is ideal for someone who learns fast, enjoys experimentation, and wants to work on cutting-edge AI applications in spacetech. Your Responsibilities Design, train, and deploy end-to-end deep-learning models (LSTM/RNN, CNN, GNN, Transformers) Define the vision, roadmap, and success metrics for AI/ML-based products Perform data cleaning, feature engineering, and exploratory data analysis to extract actionable insights Develop data pipelines and design domain-specific features or embeddings; automate feature generation and validation Prototype and productionize multi-sensor fusion workflows blending satellite and multi-source data, compressing petabytes into millisecond APIs Own the model life-cycle: data curation, experiment tracking, MLOps, CI/CD, GPU/TPU training, scalable inference on cloud Experience & Must-Have Skill Sets 2–3 yrs hands-on experience building & shipping deep-tech ML products Strong foundation in statistics, probability, and ML algorithms; solid grasp of time-series & image models (ConvLSTM, Attention, Vision Transformers) Experience with SQL, data wrangling, and large datasets Proficiency in Python, R, and libraries such as scikit-learn, TensorFlow, PyTorch Excellent communication, stakeholder management, and leadership skills Bachelor’s/Master’s degree in Computer Science, Data Science, or a related field Ownership mindset and ability to execute projects end-to-end Bonus Skills Prior experience in large-scale time-series forecasting Experience with MLOps and automation Familiarity with AWS/GCP GPU fleets, WebAssembly Startup, hackathon, or open-source contribution experience What’s in it for You Competitive compensation: high fixed salary + founding-team ESOPs Autonomy to own core feature areas end-to-end Opportunity to work on frontier deep-tech challenges Mentorship from advisors in SmartSat CRC, ISRO, IIT Delhi/Bombay, Australian Space Agency, and top global labs Flat structure, rapid experimentation, and tech-first culture About the Team The co-founders are graduates from IIT Bombay and IIM Calcutta with prior experience at Strategy&, PwC, and ISRO. How to Apply Submit your resume/CV to contact@biosky.tech Provide links to any GitHub repositories or product samples demonstrating your skill set.