Growhut
Website:
growhut.ai
Job details:
AI Engineer – Agent Systems & Persistent Memory
Company: Growhut Technologies Private Limited
Type: Full-Time
Location: Remote
Salary: Upto 5 LPA
Experience: 0-2 Years
🌱 About Growhut
Growhut is an AI-first product engineering company building production-grade systems at the intersection of AI, real-time collaboration, and intelligent automation.
We don’t build demos.
We build AI systems that run businesses.
A core focus area for us is agent-based AI—systems that reason, learn, act, and improve over time within complex, real-world workflows.
This role is for engineers who have already shipped AI to production and want to go deeper.
🎯 Role Overview
As an AI Engineer at Growhut, you will design, build, and scale AI agents that operate inside complex workflows with persistent memory, tool usage, and multi-step reasoning.
You will work on systems where:
- Conversations turn into a state
- Memory evolves over time
- Agents execute real tasks, not just generate text
This is a hands-on engineering role, not a research-only position.
🔧 Key Responsibilities
AI Agents & Workflow Orchestration
- Design and build AI agents capable of handling complex, multi-step workflows.
- Implement agent behaviors such as:
- Planning
- Tool selection
- Decision-making
- Retry and fallback logic
- Orchestrate agents across asynchronous, event-driven systems.
Persistent Memory Systems (Core Focus)
- Design and implement persistent memory architectures for AI agents.
- Work with:
- Long-term vs short-term memory separation
- Context anchoring across sessions
- Memory retrieval and relevance scoring
- Build hybrid memory systems using:
- Vector databases
- Structured stores
- Knowledge graphs (where applicable)
LLM & Retrieval Engineering
- Integrate and optimize LLMs for production use cases.
- Build and tune:
- RAG pipelines
- Graph-augmented or hybrid retrieval systems
- Optimize prompts, context windows, and cost-performance tradeoffs.
Production & Systems Engineering
- Deploy AI systems in production environments.
- Monitor latency, failure modes, hallucinations, and drift.
- Implement guardrails, logging, and observability for agent behavior.
- Collaborate closely with backend, frontend, and product teams.
📌 Required Skills
- 0-2 years of experience building AI-powered systems in production.
- Strong experience with LLM-based agents (not just chatbots).
- Hands-on experience implementing persistent memory for AI systems.
- Strong programming skills in Python (preferred) or Node.js.
- Experience with embeddings and semantic retrieval.
- Solid understanding of system design and async workflows.
🧩 Strongly Preferred (High Signal)
Experience with agent frameworks such as:
- LangChain
- LlamaIndex
- CrewAI
- Experience with vector databases (FAISS, Pinecone, Weaviate, etc.).
- Exposure to knowledge graphs or structured memory systems.
- Experience deploying AI systems on cloud platforms (AWS/GCP/Azure).
- Familiarity with evaluation, prompt versioning, and memory testing.
🧠 Who Should Apply
- You have shipped AI agents that do real work, not just generate text.
- You’ve dealt with long-term context, memory consistency, and failures.
- You enjoy designing systems, not just calling APIs.
- You’re comfortable owning production AI systems end-to-end.
🚀 Why Growhut?
- Work on serious agent-based AI systems, not experiments.
- High ownership and technical depth.
- Real production challenges across multiple domains.
- Remote-first culture with strong engineering standards.
- Opportunity to shape how AI agents are built and deployed at scale.
If you want to build AI that remembers, reasons, and executes reliably in production,
Growhut is where you should be.
Click on Apply to know more.