19 Nov
|
Asper.ai
|
Bengaluru
19 Nov
Asper.ai
Bengaluru
We are looking for a GenAI / Agentic Engineer with 4–6 years of overall experience, a solid software engineering foundation and at least 2 year of hands-on work with LLMs or agentic systems. You will build, test and ship components of enterprise-grade AI agents that power high-impact decision workflows, working closely with senior agentic engineers who own the overall architecture. You will work on tool-using agents, planning–execution–verification loops, retrieval and embeddings, context management, evaluation and observability, integrated with large datasets, enterprise data and applications.
Key Responsibilities
• Build and maintain agentic workflows (single-agent, and contributing to multi-agent setups) for business decisions and task automation, under the guidance of senior engineers
• Implement agent components that plan, reason, act, verify and recover using tools, memory and context
• Integrate LLMs (e.g., GPT-4 / GPT-4o or equivalent) with enterprise systems via APIs and event-driven services
• Develop orchestration using frameworks such as LangChain, LangGraph, Semantic Kernel or in-house agent runtimes
• Write production-quality backend code: APIs, data pipelines, unit and integration tests, and deployments
• Implement guardrails: error handling, retries, fallback logic and validation around LLM outputs
• Build and run evaluations (task success, accuracy, latency, safety) and add logging and tracing for agent behaviour
• Work on prompt engineering, retrieval (RAG), embeddings and context management to improve quality and latency
• Collaborate with product, domain SMEs and data teams to turn business logic into agent behaviour
• Follow security, access control and auditability standards for agent actions and data Requirements
• 4–6 years of overall experience in software engineering, with at least 2 year hands-on in GenAI / LLM applications (production, internship or substantial personal/open-source projects)
• Hands-on experience building LLM-based or agentic applications (tool calling, RAG, memory, routing or planning)
• Strong Python skills (C++ is a plus), with good backend practices: REST APIs, async code, testing, Git, code reviews
• Working knowledge of prompt engineering, embeddings, vector databases and context management
• Exposure to at least one orchestration or evaluation framework (e.g., LangChain, LangGraph, LangSmith, TruLens, RAGAS)
• Comfort working with APIs and data transformation (SQL, pandas or similar) across large datasets
• Basic production awareness: logging, monitoring, error handling, containers (Docker) and CI/CD
• Strong problem-solving, eagerness to learn, and clear communication Good to Have
• Exposure to multi-agent patterns (agent-to-agent communication, supervisors, control planes)
• Familiarity with Azure AI stack (Azure OpenAI, AI Search) or other cloud-native deployments
• Interest or experience in CPG, retail, pricing, RGM, finance or other data-heavy domains
• Exposure to fine-tuning (LoRA, QLoRA) or eval-driven iteration
• GitHub portfolio, hackathons, publications or open-source contributions in GenAI Education
• Bachelor's degree in Computer Science, Information Technology or a related field is a must. What Success Looks Like
• Agent features you build work reliably in production, not just in demos
• Clean, tested, well-documented code that seniors can build on
• Measurable gains in eval scores, latency and task success on the components you own
• Growing ownership: from well-scoped tasks to owning a full agent workflow within 12–18 months Why work with us?
• A fast-paced, growing and reliable start-up with a culture of care, where you are enabled to be your best personally and professionally
• Be part of the AI revolution and build solutions that make a real difference
• Work and learn alongside some of the brightest and nicest colleagues; we work hard and have fun together
📌 GenAI Engineer (Bengaluru)
🏢 Asper.ai
📍 Bengaluru