jobs in the wild.

Here’s the engineering problem. Take 20 CCTV streams from different manufacturers. Different resolutions. Unreliable networks. Poor lighting. Limited compute. Now build a system that can continuously process those streams, detect meaningful events, escalate complex observations to stronger vision models, and operate reliably at a commercially viable cost. That’s one of the problems we’re solving at NavSecura Technologies. We’re developing a real-time video intelligence system that connects existing camera infrastructure with NavArgus, our operational intelligence platform. What you’ll own: * Real-time video ingestion and processing — RTSP, FFmpeg, GStreamer. * Object detection, tracking, temporal event analysis and model fine-tuning. * Multi-stage inference using lightweight detectors and vision-language models. * Model evaluation, false-alarm reduction and dataset development. * CPU/GPU optimisation, throughput, latency and inference economics. * Production reliability, observability and integration with NavArgus. Who we’re looking for: An engineer with strong Python, PyTorch and computer vision fundamentals who has built and tested working systems. Experience deploying video models, handling continuous streams or optimising inference is particularly valuable. We care more about demonstrated engineering ability than the number of frameworks on your résumé. This is a founding engineering position. You’ll have direct technical ownership, work closely with the founder and help make foundational architecture decisions. The backend foundation is under development. The production vision system is still to be built and validated. Location: India — Remote-first Employment: Full-time Role: Founding Computer Vision Engineer Salary: ₹10–14 LPA (annual CTC) To apply: Send your résumé and one computer vision project you’ve personally built. Tell us three things: 1. What was the technical problem? 2. What did you personally build? 3. What failed, and how did you fix it? NavSecura Technologies #ComputerVision #Hiring #PyTorch #VideoAI #FoundingEngineer

📍 Remote | 🕒 5–8+ yrs | 💼 Full-time We’re looking for a hands-on engineer with deep graph database expertise to architect and optimize large-scale knowledge graph systems powering enterprise GenAI applications. What you’ll do: ▪ Build scalable knowledge graphs: entity extraction, relationship inference, entity linking, semantic enrichment, at billions of nodes ▪ Design graph-backed RAG systems and multi-hop reasoning ▪ Optimize Neo4j performance: Cypher, query planning, b-tree/full-text/HNSW indexing, traversal and caching ▪ Build distributed ingest pipelines with idempotence, state management, consistency and monitoring ▪ Evaluate and integrate graph DBs, vector stores, semantic search and LLM APIs What you bring: ‣ Neo4j, graph data modeling, and graph algorithms (PageRank, centrality, community detection, k-NN, link prediction) ‣ Data engineering: ETL/ELT, data lineage and quality, structured and unstructured data at scale ‣ SQL, NoSQL, and vector databases (ANN, HNSW) ‣ Azure: Functions, AKS, CosmosDB, Durable Functions, Service Bus, Event Hub, Event Grid, Synapse, Data Factory, Cognitive Search, Key Vault, RBAC ‣ IaC (Bicep, Terraform, ARM), distributed systems, event-driven and async architectures, microservices ‣ Python/Java, API design, observability (App Insights, ELK, Prometheus) ‣ GenAI: OpenAI, Azure OpenAI, Claude, prompt engineering, RAG, embeddings, inference and cost optimization ‣ 2+ yrs building Neo4j knowledge graphs and 2+ yrs deploying GenAI systems ‣ Master’s in CS, Engineering or Math (Ph.D. preferred) Nice to have: › Neo4j or Azure certifications (AZ-104, AZ-305, DP-900) › GNNs (PyTorch Geometric, DGL, GraphSAGE, Node2Vec), RDF/semantic web › Spark GraphX, Gremlin, sharding, and multi-tenancy at 100M+ nodes › Docker, Kubernetes, Helm › Open-source contributions 📩 Apply now or DM me | Refer someone great! 👇 #Hiring #KnowledgeGraph #Neo4j #GraphDatabase #GenAI #RAG