Who we are
Xenoss is an AI engineering and integration services company, helping medium to large enterprises run AI transformation end-to-end, from situation analysis and goals framing to data discovery and preparation, pipeline building, model development, retraining pipeline design, solution deployment, and support. We build a broad spectrum of AI solutions such as user behaviour prediction, content generation, NLP, audience segmentation, pathfinding solutions, AI assistants, edge computer vision, fraud detection, and others. We work with prominent companies such as Microsoft, Toshiba, AstraZeneca, Activision Blizzard, Verve Group, Voodoo Games, and Telefonica, among others. We’re included in the top 100 software companies on the Inc. 5000 list.
Role description
We’re looking for a practical builder, not an ML researcher or a prompt engineer. You’ll work directly with technical leadership on ambiguous, fast-moving projects: integrating LLMs, wiring up data pipelines, shipping agentic workflows, connecting external APIs and tools, and demonstrating end-to-end value quickly. Success in this role looks like trusted execution. We can hand you a problem ,statement and expect a working prototype, clear trade-offs, and code we’d be comfortable extending, while you grow your system design and product judgment through mentorship.
What will you do
You will build end-to-end AI-enabled product features, prototypes, and internal tools across our client engagements.
Core work includes: ● Applied AI delivery: Design and implement LLM-powered features: RAG, tool-using agents, eval hooks, and prompt/context patterns. You ship to staging- and production- oriented quality, not notebook demos. ● Backend & data: Build Python services, APIs, and background jobs with SQL/Supabase- style data access, ingestion, and retrieval pipelines. You keep schemas sensible and logging in place. ● Integrations: Connect MCP, REST, webhooks, and third-party APIs (e.g. Composio, Supabase, email and calendar patterns), handling auth, retries, and failure modes properly. ● Prototype full-stack: When needed, build a simple, clear demo UI (React / Next or equivalent) to prove out a flow without owning a large frontend codebase. ● AI-native SDLC: Use Cursor / Claude (or equivalent) daily for implementation, tests, refactors, and docs, and orchestrate agent-assisted workflows (skills, hooks, multi-step tasks) with human review at merge boundaries. ● Engineering judgment: Bring strong programming fundamentals (types, async, testing, debugging, Git). You find solutions quickly without sacrificing maintainability, and you rewrite low-quality AI output before it ships. ● Ambiguity & growth: Operate with incomplete specs, document your assumptions, raise architecture questions early, and grow toward owning solution design for a feature area.
Technology landscape
You will work across a modern fullstack and applied AI engineering stack, including: ● AI-paired backend and frontend development (e.g. Claude Code, Cursor) ● Python and/or TypeScript ● LLM APIs such as OpenAI, Anthropic, Gemini, and similar platforms ● Agentic frameworks and orchestration tools such as LangGraph, LangChain, CrewAI, AutoGen, or similar ● RAG pipelines, embeddings, vector databases, and hybrid search ● Tool calling, structured outputs, workflow orchestration, and state management ● SQL, data transformations, ETL/ELT basics, and practical data handling ● Docker, cloud environments, CI/CD basics, logging, and monitoring
What should you bring
Experience ● 3–6 years of software engineering (or equivalent proven delivery), with at least 1 year of hands-on applied AI/LLM integration in real projects.
Technical ● Python: production services, async, packaging, testing (pytest). ● Backend: REST APIs, auth basics, job queues or workers, observability basics. ● Data: SQL, ETL/light pipelines, embeddings retrieval, chunking/indexing for RAG. ● Applied AI: LLM APIs (OpenAI/Anthropic/Google or similar)
