Overview

Soliton is a high-technology software company working with global customers across Semiconductor, Medical, Automotive, Industry 4.0, and High-Tech domains.

Responsibilities

• Design, implement, andoptimizeGenerative AI applicationsusing Python and frameworks such asFastAPI.
• Build AI solutions usingLLM frameworkslikeLlamaIndexandLangChain.
• Implementcontainerized deploymentsusing Docker.
• Develop andoptimizeRetrieval-Augmented Generation (RAG)pipelines for improved information retrieval.
• Work withself-hosted and cloud-based vector databasesfor efficient search and retrieval.
• Design and manageknowledgegraphs and graph-based RAG systems.
• Implementre-ranking models and retrieval optimization techniques.
• Applyprompt engineering and context engineeringto enhance model performance.
• Establishguardrailsto ensure safe, ethical, and compliant AI deployments.
• Builddata preprocessing and transformation pipelinesfor structured and unstructured data.
• Perform inference usingoffline LLMsvia platforms like Ollama or Hugging Face (Llama, Mistral).
• Integrateonline LLM providerssuch as OpenAI, Anthropic, or GCP for real-time inference.
• Monitor AI workflows usingobservability toolslikeMLfloworArizePhoenix.
• Evaluate model performance usingframeworks such asTruLensor custom-built evaluation systems.
• Continuously improve AI systems based onevaluationinsights, metrics, and user feedback.
• Experience buildingGenerative AI applicationsusing Python andFastAPI.
• Hands-on knowledge ofLLM frameworkssuch asLangChainorLlamaIndex.
• Ability to work withunstructured data(PDFs, documents, chunking, search) and structured data.
• Experience designingRAG-based systems, including prompt engineering and retrieval optimization.
• Familiarity withvector databases(Qdrant, Pinecone,Weaviate) and search solutions.
• Exposure toAI agents, workflows, and basic orchestration concepts.
• Experience usingcloud platformslike Azure or AWS.
• Working knowledge ofonline and offline LLMs(OpenAI, Llama, Mistral).
• Understanding ofAI evaluation, monitoring, and observabilityconcepts.
• Experience withDocker and CI/CD pipelinesfor deploying AI applications.
• Experience with MCP clients and servers
• Knowledge of multimodal LLMs for image and voice processing
• Knowledge of deploying applications in cloud or on-prem infrastructure
• Knowledge of fine-tuning techniques and data preparation for fine-tuning

Qualifications

• Bachelor’s orMaster’s degree in Computer Science, Engineering, or a related field
• Proven experience in AI/ML engineering and related technologies
• 3+ years of experience building applications using Python and asynchronous programming
• Experience working with SQL and NoSQL databases
• Strong problem-solving skills and ability to work in a fast-paced environment
• Excellent communication and teamwork skills

Benefits

• We want every Soliton member to grow to their highest potential. Our work environment helps individuals explore their interests and potential and reach out to the resources and people available at Soliton to realize them. Read more about whatit’slike to work at Soliton.Soliton Employee Value Proposition.
• Solitons choose their work hours as long asthey takeinto account the requirements of the job.We takespecial careto support mothers to excel at work while they handle their responsibilities at home.
• At Soliton, we believe that every team member contributes to our success and revenue, directly or indirectly. To recognize this, we share aportionof our profits with all Solitons. Starting from your second year with us,you’llbe eligible to receive a share of the company’s profits.
• Health insurance for employees and families,gymand cycle allowance – your health is a priority!
• Work Location (Bangalore/Coimbatore) :This role will require working from the office (WFO) for the first 12 months. Based on individual performance and business requirements, a remote or hybrid workoptionmay be considered after one year.
• For more information, visithttps://www.solitontech.com/anddo readtheImpact Reportto get a glimpse of the first 25 years of our truly meaningful journey.