Automation and AI Developer
Thomson Reuters3 days ago
Hyderābād, IndiaMid Level / Senior
Responsibilities
- Translate current business processes and SME knowledge into buildable AI requirements.
- Support feasibility studies and proof-of-concept development with architects and stakeholders.
- Design and implement complex AI solution components end-to-end on AWS AgentCore and LangGraph.
- Scale AI solutions from proof of concept to production, including performance, reliability, monitoring, and cost management.
- Debug production issues involving hallucinations, retrieval degradation, agent loops, and latency regressions.
- Develop low-code automations with Copilot Studio and Power Automate when appropriate.
- Make day-to-day technical decisions, identify architectural risks, and translate architecture into implementation plans.
- Guide developers and associates through implementation, code reviews, and debugging.
- Demo solutions to internal stakeholders and maintain communication through delivery.
Requirements
- 3–5 years of total experience, including 2–4 years of production GenAI/LLM experience.
- 12–24 months of multi-agent LangGraph experience and ownership of at least three systems after deployment.
- Hands-on expertise with LangGraph multi-agent state machines, parallel execution, checkpointing, and human-in-the-loop workflows.
- Experience with AWS Bedrock, AWS AgentCore, model deployment, knowledge bases, and guardrails.
- SQL proficiency, NL-to-SQL, LLM-powered query layers, and familiarity with Snowflake as a plus.
- Experience developing REST APIs, using enterprise system connectors, and implementing MCP and A2A servers or clients in production.
- Knowledge of advanced RAG, hybrid search, re-ranking, query reformulation, and retrieval evaluation.
- Experience with RAGAS or TruLens, golden test sets, automated regression pipelines, and LangSmith or LangFuse observability.
- Expert Python skills, including design patterns, asynchronous programming, testing, and CI/CD.
- Experience with OpenAI, Claude, Gemini, and Llama models and production-scale model trade-offs.
- Working knowledge of Copilot Studio, Power Automate, Claude Code, GitHub Copilot, Cursor, or Cline.
- Experience with document intelligence, OCR, layout-aware chunking, and LoRA or QLoRA fine-tuning.
- Ability to communicate with stakeholders, demonstrate AI solutions, and translate business needs into technical requirements.
- AWS ML Specialty, Azure AI Engineer, or GCP ML Engineer certification is relevant.
Benefits
- Hybrid work model with three days from the office.
- Work-from-anywhere flexibility for up to eight weeks per year, subject to company policies.
- Flexible workplace policies supporting work-life balance and personal responsibilities.
- Career development, continuous learning, and skills-development programs.
- Flexible vacation, two company-wide Mental Health Days, Headspace access, retirement savings, tuition reimbursement, incentive programs, and wellbeing resources.
- Two paid volunteer days annually and opportunities for pro-bono consulting and ESG initiatives.
- Shift timing is 1pm–10pm IST.
Tech Stack
Categories
About Thomson Reuters
Thomson Reuters builds research platforms, workflow software, and data services for legal, tax and accounting, compliance, and government professionals, and operates the Reuters global news service. Flagship products include Westlaw for legal research and ONESOURCE and Checkpoint for tax and accounting, sold primarily via subscriptions and enterprise licenses. A public company headquartered in Toronto, it was formed in 2008 by combining Thomson Corporation and Reuters Group and is listed on the TSX and Nasdaq as TRI.