AI Solutions Engineer

DhanInfo is looking for an AI Solutions Engineer to design and build practical AI-powered solutions for real-world business and operational problems.

We work in a BPO environment where projects can vary significantly across clients and processes. One project may involve analyzing thousands of customer calls, another may require processing images or documents, while others may focus on workflow automation, AI agents, data extraction, or improving the efficiency of an existing process.

This is a hands-on role for someone who can move from:

Business Problem Solution Design Prototype Development Integration Production

We are looking for someone who understands AI deeply enough to choose the right approach, but is equally comfortable building, integrating, debugging, and deploying the solution.

What Youll Work On

Projects may span areas such as:

Voice & Conversation AI

• Speech-to-text and call transcription

• Call classification and summarization

• Extracting structured information from conversations

• Automated QA and compliance checks

• Sentiment and conversation analysis

• Voice agents

• Integrating call insights into CRMs and internal systems

Image & Document Processing

• OCR and document understanding

• Image classification and object detection

• Processing scanned documents, forms, and photographs

• Multimodal AI workflows

• Automated verification and validation processes

Generative AI & LLM Applications

• AI assistants and copilots

• Retrieval-Augmented Generation (RAG)

• AI agents and tool calling

• Structured data extraction

• Classification and summarization

• Knowledge search

• LLM evaluation and hallucination reduction

Process Automation & Efficiency

• Understanding existing BPO workflows

• Identifying opportunities for automation

• Reducing repetitive manual work

• Connecting systems through APIs and webhooks

• Human-in-the-loop workflows

• Intelligent routing and decision-making

• Automated data entry and validation

• Improving turnaround time, accuracy, and operational efficiency

Key Responsibilities

• Work with business and operations teams to understand existing processes and pain points.

• Identify where AI, automation, or traditional software can create measurable value.

• Translate business requirements into technical solutions and system architectures.

• Build rapid proof-of-concepts to validate ideas before full-scale development.

• Develop backend services and AI applications primarily using Python.

• Integrate LLMs, speech models, vision models, APIs, databases, and third-party systems.

• Build extraction, classification, summarization, and decision-making pipelines.

• Design AI workflows with appropriate validation, confidence thresholds, and fallback mechanisms.

• Implement human-in-the-loop processes where full automation is not reliable or appropriate.

• Build and consume REST APIs, webhooks, and external integrations.

• Evaluate models and approaches based on accuracy, latency, reliability, and cost.

• Deploy, monitor, troubleshoot, and continuously improve production AI systems.

Measure business impact such as:

• Manual hours saved

• Automation percentage

• Reduction in turnaround time

• Accuracy improvement

• Cost per transaction

• Communicate technical solutions clearly to both technical and non-technical stakeholders.

• Take ownership of projects from initial discovery through production deployment.

Required Experience & Skills

5+ years of overall experience in software engineering, AI/ML, automation, data products, or related technical roles.

Strong recent hands-on experience building AI-powered applications or automation solutions.

Strong programming skills in Python.

Experience building backend applications or APIs using frameworks such as:

• FastAPI

• Flask

• Django

Hands-on experience with LLM platforms such as:

• OpenAI / Azure OpenAI

• Claude

• Gemini

• Open-source LLMs

Strong understanding of:

• Prompt engineering

• Structured outputs

• RAG

• Embeddings

• AI agents

• Tool/function calling

• LLM evaluation

• Hallucination mitigation

Experience working with REST APIs, webhooks, and third-party integrations.

Working knowledge of relational and/or NoSQL databases.

Strong debugging and problem-solving ability.

Ability to independently take an ambiguous business problem and turn it into a working technical solution.

Ability to evaluate whether a problem should be solved using AI, automation, deterministic logic, or a combination of approaches.

Good to Have

Experience with any of the following is useful, but candidates are not expected to know every tool or technology listed:

• Whisper, Deepgram, AssemblyAI, or similar speech platforms

• Vapi, Retell, ElevenLabs, or other Voice AI platforms

• OCR and Document AI

• OpenCV, YOLO, or other computer vision technologies

• Multimodal LLMs

• LangChain, LangGraph, LlamaIndex, or similar frameworks

• Pinecone, Qdrant, Weaviate, pgvector, or other vector databases

• n8n, Make, Zapier, or workflow automation platforms

• Browser automation

• Docker and containerized deployments

• AWS, Azure, or GCP

• Redis, queues, and asynchronous processing

• CI/CD pipelines

• High-volume data or transaction-processing systems

What Were Looking For

We are not looking for someone who simply knows AI terminology or has worked with a particular framework.

We are looking for someone who can be given a problem such as:

Our team manually listens to thousands of calls, extracts information, checks whether agents followed the correct process, and updates multiple systems.

and determine:

• Which parts can realistically be automated

• Which AI or software approach should be used

• How should the workflow be architected

• Where should deterministic rules be used instead of an LLM

• How should accuracy be measured

• What happens when AI confidence is low

• Where should human review be introduced

• How should the solution integrate with existing systems

• How much will it cost to operate

• What measurable efficiency improvement will it create

The next project could involve images, documents, browser automation, operational data, or a completely different workflow.

The ability to learn quickly, understand unfamiliar processes, and select the right solution is therefore critical.

Ideal Candidate

You will likely be successful in this role if you:

• Enjoy solving ambiguous and practical business problems.

• Are hands-on and comfortable building solutions yourself.

• Can communicate effectively with both operations teams and developers.

• Understand both the capabilities and limitations of AI.

• Know when traditional software or rules are better than an LLM.

• Think about reliability, edge cases, and failure scenarios.

• Consider accuracy, latency, and cost when designing systems.

• Can prototype quickly and iterate based on real-world feedback.

• Are comfortable working across multiple projects and AI domains.

• Focus on business outcomes rather than using AI for its own sake.

Success in This Role

Success will not be measured by the number of AI tools or frameworks used. It will be measured by whether the solutions you build:

• Reduce manual work

• Improve operational accuracy

• Increase process efficiency

• Reduce turnaround time

• Integrate reliably into existing workflows

• Scale to real production usage

• Deliver measurable business value

This role offers the opportunity to work across Generative AI, Voice AI, Computer Vision, Automation, and business process transformation while solving real operational problems at scale

Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.