Associate Developer
Responsibilities
Prompt & Retrieval Engineering
- Design dynamic prompt templates for diverse enterprise use cases (support, sales, operations, HR).
- Build context-aware RAG pipelines using structured (SQL) and unstructured (PDFs, CSVs) sources.
- Tune prompts to improve response precision, latency, tone, and relevance for chat and voice channels.
Multi-Agent LLM System Design
- Architect and implement agents for tasks like entity extraction, table/column identification, reasoning, and response generation.
- Enable tool use and API integration within LLM calls (e.g., function/tool calling via Bedrock or OpenAI APIs).
- Support deterministic prompt generation and fallback flows.
Voice Agent Integration
- Work with our voice stack (Deepgram, WebRTC, Amazon Polly, Twilio) to optimize LLM prompts for speech.
- Design effective real-time strategies for transcription correction (e.g., brand name phonetics) and TTS personalization.
Performance Monitoring & Feedback Loops
- Instrument AI pipelines to track performance metrics (accuracy, latency, user drop-off).
- Collaborate with internal teams to use user feedback and logs to auto-tune prompts and context injection flows.
- Contribute to the
feedback validation system
in Smartbots Studio.
Platform & MLOps Enablement
- Work with backend teams (Python/Flask, Node.js, Lambda) to integrate AI modules into our orchestrator layer.
- Leverage Redis, Amazon RDS, and Bedrock-based workflows to support high-scale, low-latency interactions.
R&D and Thought Leadership
- Stay up-to-date with the latest in GenAI (Claude 3.5, GPT-4o, open-weight models), vector databases, and knowledge graphs.
- Evaluate new models, fine-tuning approaches, or hybrid pipelines based on specific enterprise problems.
- Contribute to internal playbooks and reusable prompt libraries.
What You’ll Bring
Hands-on experience
with prompt engineering and RAG workflows across real-world deployments.- Experience with
LLMs
like GPT, Claude, LLaMA, Mistral, or Phi. - Proficient in
Python
(for prompt pipelines, embedding generation, API orchestration). - Comfortable with
microservice integration
(Flask, Node.js), Redis caching, and REST API development. - Working knowledge of
React or Next.js
is a plus (our Studio is built on it). - Experience designing for
real-time voice
experiences is a strong advantage. - Familiarity with
AWS Bedrock, Amazon Transcribe/Polly, Deepgram
, or similar stacks. - Strong documentation and communication skills—able to write clear instructions and debug prompt chains.
Why Join Smartbots
Real Enterprise Impact
: Power assistants used by top companies in telecom, insurance, pharma, and semiconductors.Platform-First Approach
: Work on Smartbots Studio, our flagship no-code builder for chat and voice agents.R&D Culture
: Tinker with multi-agent architectures, custom embeddings, and domain-specific modeling.Collaborative Team
: Work alongside GenAI engineers, backend developers, and UX designers.Flexible Working
: Hybrid or remote options with in-person team jams at our Hyderabad HQ.Competitive Compensation
: Salary, performance incentives, and upskilling budgets.
About Smartbots.ai
Smartbots is a fast-growing conversational AI company helping enterprises transform customer support, internal operations, and data intelligence using LLM-based AI agents. Our platform empowers teams to rapidly build, test, and scale chatbots, voicebots, and BI assistants with deep context and real-time data integration.
