Full Stack Engineer

About Valiance

Valiance is a 150+ person AI products and solutions company building enterprise-grade AI systems for clients across industrial, government, and global enterprise segments. We're a Google Cloud Premier AI Partner, and our work has been recognized with the Nasscom AI GameChanger Award and the Aegis Graham Bell Award for Innovation in AI

We build two flagship products — Tender Intelligence (AI-powered procurement monitoring) and Vision Intelligence (computer vision for industrial safety and security) — alongside bespoke agentic AI deployments for enterprise and public sector clients. Our engineers work close to the metal: shipping production agents, RAG pipelines, and multimodal AI systems, not just prototypes.

About the Role

We're hiring a Software Engineer to build agentic AI applications — systems where LLMs don't just answer questions but plan, call tools, retrieve context, and take actions inside real business workflows. You'll work across the stack: frontend, backend, and cloud deployment, and you'll be expected to use AI-assisted coding agents (Claude Code or similar) as a core part of how you write and ship software — not as a novelty, but as a daily multiplier on your outp

ut.This is a hands-on builder role. You'll move fast from spec to working product, often translating a functional requirements doc directly into a deployable application with AI-assisted development workflows.

What You'll Do

• Design and build agentic AI applications — multi-step LLM workflows involving tool use, function calling, memory, and orchestration (e.g., LangGraph, custom agent loops, or similar frameworks)
• Build RAG pipelines — document ingestion, chunking, embeddings, vector search, and grounded generation — for use cases like field intelligence agents, technical document Q&A, and compliance copilots.
• Develop full-stack features: REST/GraphQL APIs and backend services (Python/Node.js) plus frontend interfaces (React/Next.js) for internal and client-facing AI products.
• Deploy and operate applications on Google Cloud Platform — Cloud Run, GKE, Vertex AI, Cloud SQL/Firestore, Pub/Sub — with attention to cost, latency, and scalability.
• Use AI-assisted coding agents (Claude Code or equivalent) as a primary development tool — writing specs/prompts that let the agent scaffold, refactor, and ship code, while you own architecture, code review, and correctness.
• Integrate with third-party and internal APIs/MCP servers (e.g., Slack, Google Workspace, CRM systems, client-specific data sources) to extend agent capabilities.
• Work directly with functional specs from product/pre-sales and convert them into working software with minimal hand-holing
• Write tests, set up CI/CD pipelines, and maintain production systems post-launch (monitoring, logging, incident respose)
• Collaborate with solution architects and the founder/CEO on technical feasibility during pre-sales and client scoping calls when needed.

Must-Have Skills

• 2–5 years of professional software engineering experience.
• Hands-on experience building at least one production or near-production AI/agentic application — RAG system, LLM-powered chatbot/copilot, or autonomous agent (not just a tutorial project)
• Strong backend skills in Python (FastAPI/Flask/Django) — Node.js is a plus.
• Working frontend skills in React/Next.js — you don't need to be a design expert, but you should ship usable, clean UIs without a dedicated frontend engineer.
• Practical experience with LLM APIs (Anthropic Claude, OpenAI, Gemini) — prompt design, function/tool calling, structured outputs, streaming.
• Experience with at least one agent framework or pattern — LangChain/LangGraph, custom orchestration, or comparable.
• Familiarity with vector databases (Pinecone, Weaviate, pgvector, or similar) and embedding-based retrieval
• Real, regular usage of AI-assisted coding tools (Claude Code, Cursor, Copilot, etc.) in your actual development workflow — you should be comfortable directing an AI agent to write/refactor code and critically reviewing its output.
• Experience deploying applications on a major cloud provider — GCP strongly preferred (Cloud Run, GKE, IAM, Vertex AI); AWS/Azure experience is transferable.
• Comfort with Docker and basic CI/CD (GitHub Actions or equivalent)
• Solid fundamentals: REST API design, SQL/NoSQL databases, Git workfows.

Nice to

• HaveExperience with computer vision (OpenCV, YOLO, or similar) — relevant to our Vision Intelligence productline.
• Exposure to Google Vertex AI / Agentspace / Gemini Enterprise.
• Experience working in a client-facing or consulting environment (translating ambiguous requirements into shippable specs)
• Prior startup or small-team experience where you owned a feature end-to-end.
• Open-source contributions to AI/agent tooling.

What We're Looking For (Beyond the Checklist)

• You move fast and you're not afraid to ship — but you also know when to slow down on architecture decisions that are hard to reverse.
• You treat AI coding agents as a force multiplier, not a crutch — you can read and debug code an agent wrote just as well as code you wrote yourself.
• You're comfortable with ambiguity — specs from clients aren't always complete, and you can ask the right questions or make sound assumptions.
• You want to work on real production AI systems used by large enterprises and government bodies, not just internal demos.

What We Offer

• Direct exposure to enterprise and public-sector AI deployments at scale.
• Work alongside a tight, technical leadership team — fast decision cycles, no bureaucracy.
• Access to cutting-edge tooling: latest Claude models, Claude Code, GCP credits/partner resources including discounted GPU compute (H200, RTX Pro 6000)
• Competitive compensation, performance-linked growth.
• A flat, builder-first culture where shipped product speaks louder than tenure.