Headline

Software Engineer - AI @ Barclays | Backend & AI Engineer | Java | Spring Boot | Microservices | Distributed Systems | AWS | Agents, LLMs & Prompt Engineering | Problem Solving | Research @ Harvard, CMU & Stanford

About

Software Engineer at Barclays, specializing in backend systems, distributed architectures, and enterprise-grade AI automation. I work at the intersection of Java backend engineering and agentic AI, building platforms that automate complex engineering workflows and support large-scale financial systems. ▸ Backend & Platform Engineering: Building scalable services using 𝗝𝗮𝘃𝗮, 𝗣𝘆𝘁𝗵𝗼𝗻, 𝗮𝗻𝗱 𝗦𝗽𝗿𝗶𝗻𝗴 𝗕𝗼𝗼𝘁, with distributed data pipelines, ETL, Kafka, Apache Spark, and SQL. ▸ Agentic AI & Enterprise Automation: Developing 𝗟𝗟𝗠 𝗮𝗻𝗱 𝗥𝗔𝗚-powered systems and agentic workflows for migration, testing, release, and deployment automation, reducing manual oversight by 70%. ▸ Enterprise AI Platforms: Building 𝗡𝗟𝗣 𝗮𝗻𝗱 𝗥𝗔𝗚-based engineering assistants integrated with Jira, GitLab, and Confluence through custom 𝗠𝗖𝗣 𝘀𝗲𝗿𝘃𝗲𝗿𝘀, achieving 95%+ accuracy on complex engineering queries. ▸ Distributed Systems & Performance: Working with 𝗔𝗽𝗮𝗰𝗵𝗲 𝗜𝗴𝗻𝗶𝘁𝗲 and high-throughput financial systems, improving data access and reducing latency by up to 30%. ▸ AI Research & Collaboration: Conducting research targeting top-tier AI conferences and workshops, including 𝐂𝐕𝐏𝐑, 𝐈𝐂𝐂𝐕, 𝐄𝐂𝐂𝐕, 𝐍𝐞𝐮𝐫𝐈𝐏𝐒, 𝐈𝐂𝐋𝐑, 𝐚𝐧𝐝 𝐀𝐀𝐀𝐈, 𝐚𝐜𝐫𝐨𝐬𝐬 𝐮𝐧𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲 𝐢𝐧 𝐀𝐈, 𝐛𝐞𝐧𝐜𝐡𝐦𝐚𝐫𝐤𝐢𝐧𝐠, 𝐚𝐠𝐞𝐧𝐭𝐢𝐜 𝐀𝐈, 𝐕𝐋𝐌𝐬, 𝐦𝐮𝐥𝐭𝐢𝐦𝐨𝐝𝐚𝐥 𝐥𝐞𝐚𝐫𝐧𝐢𝐧𝐠, 𝐚𝐧𝐝 𝐫𝐞𝐚𝐬𝐨𝐧𝐢𝐧𝐠 𝐬𝐲𝐬𝐭𝐞𝐦𝐬. Collaborating with researchers across academia and industry while contributing to open-source AI systems, datasets, benchmarks, and evaluation frameworks. ▸ Quantitative & Low-Latency Systems: Building 𝗖++ 𝗹𝗼𝘄-𝗹𝗮𝘁𝗲𝗻𝗰𝘆 𝘀𝘆𝘀𝘁𝗲𝗺𝘀 and quantitative research tooling, including a limit order book and matching engine achieving ~3M orders/sec and a Python/C++ statistical backtesting framework. 𝐒𝐤𝐢𝐥𝐥𝐬: 𝐉𝐚𝐯𝐚 · 𝐏𝐲𝐭𝐡𝐨𝐧 · 𝐂++ · 𝐒𝐩𝐫𝐢𝐧𝐠 𝐁𝐨𝐨𝐭 · 𝐊𝐚𝐟𝐤𝐚 · 𝐀𝐩𝐚𝐜𝐡𝐞 𝐒𝐩𝐚𝐫𝐤 · 𝐒𝐐𝐋 · 𝐀𝐖𝐒 · 𝐃𝐨𝐜𝐤𝐞𝐫 · 𝐊𝐮𝐛𝐞𝐫𝐧𝐞𝐭𝐞𝐬 · 𝐋𝐋𝐌𝐬 · 𝐀𝐠𝐞𝐧𝐭𝐢𝐜 𝐀𝐈 · 𝐕𝐋𝐌𝐬 · 𝐏𝐲𝐓𝐨𝐫𝐜𝐡 · 𝐀𝐈/𝐌𝐋 · 𝗡𝗟𝗣 · 𝗥𝗔𝗚 · 𝗠𝗖𝗣 📩 Open to Software Engineering, AI/ML, Quantitative Engineering roles, research collaborations, and open-source contributions.

Location

Pune District

AI stack

antigravity

claude code

cursor

geminicli

grok

hugging face

Work experience

Research Project Lead at Harvard Medical School, Apr 2026 - Present · 6 mos, Remote, • Led a collaborative research project on unsupervised 𝗱𝗲𝗲𝗽 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 for neuroimaging, focusing on image registration and segmentation across diverse imaging modalities. • Developed PyTorch-based diffusion and geometric-matching frameworks for 𝗠𝗥𝗜, 𝗢𝗖𝗧, 𝗮𝗻𝗱 𝗟𝗦𝗙𝗠 neuroimaging datasets. • Coordinated research experiments, technical contributions, and project milestones toward a journal-targeted publication

Researcher at Stanford University, Mar 2026 - Present · 7 mos, Remote, • Research reasoning and 𝘂𝗻𝗰𝗲𝗿𝘁𝗮𝗶𝗻𝘁𝘆 𝗶𝗻 𝗔𝗜, investigating how LLMs represent, quantify, and reason under uncertainty through DA-LLM research. • Design experiments and evaluation methodologies to assess reasoning reliability and model behavior under ambiguous and uncertain conditions. • Work across 𝗟𝗟𝗠𝘀, 𝗿𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴 𝘀𝘆𝘀𝘁𝗲𝗺𝘀, 𝗮𝗻𝗱 𝗔𝗜 𝗲𝘃𝗮𝗹𝘂𝗮𝘁𝗶𝗼𝗻 𝗳𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸𝘀, developing experimental approaches to study model reliability and decision-making.

Research Intern at Carnegie Mellon University, Jan 2026 - Present · 9 mos, Remote, Contributed to research on 𝗮𝗴𝗲𝗻𝘁𝗶𝗰 𝗔𝗜, 𝗟𝗟𝗠𝘀, 𝗩𝗟𝗠𝘀, 𝗺𝘂𝗹𝘁𝗶𝗺𝗼𝗱𝗮𝗹 𝗿𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴, 𝗮𝗻𝗱 𝗔𝗜 𝗿𝗲𝗹𝗶𝗮𝗯𝗶𝗹𝗶𝘁𝘆, collaborating with Carnegie Mellon University–affiliated research labs on projects connecting AI systems research with real-world applications. • Led 𝗔𝗣𝗜𝗦𝗛𝗜𝗙𝗧 𝗮𝗻𝗱 𝗧𝗥𝗔𝗖𝗘, developing verifiable agent-learning and RL environments for API contract migration and autonomous network troubleshooting, respectively. • Contributed to 𝗖𝗥𝗬𝗢-𝗔𝗚𝗘𝗡𝗧, a physics-guided agentic Cryo-ET reconstruction system, and 𝗧𝗮𝗯𝘂𝗟𝗠 & 𝗞𝗶𝗻𝘆𝗮𝗘𝗺𝗯𝗲𝗱, extending language-model capabilities to structured/tabular data and low-resource language representation learning. • Built evaluation and benchmarking frameworks to measure reasoning, reliability, and agent performance across research environments, 𝗩𝗟𝗠𝘀, 𝗮𝗻𝗱 𝗟𝗟𝗠-𝗯𝗮𝘀𝗲𝗱 systems. Currently exploring SR-LoRA and Agentic VAD, investigating parameter-efficient fine-tuning and agentic reasoning. Research work has been submitted to venues including 𝗘𝗖𝗖𝗩, 𝗔𝗔𝗔𝗜, 𝗖𝗩𝗣𝗥, 𝗮𝗻𝗱 𝗡𝗲𝘂𝗿𝗜𝗣𝗦...

Software Engineer at Barclays, Jul 2025 - Present · 1 yr 3 mos, Pune District

AI Trainer – Generative AI Specialist at Geeksoft LLC, Mar 2025 - May 2025 · 3 mos, Remote, • Delivered specialized Generative AI training to experienced professionals, covering 𝗟𝗟𝗠𝘀, 𝗧𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺𝗲𝗿𝘀, 𝗥𝗡𝗡𝘀, 𝗟𝗦𝗧𝗠𝘀, 𝗚𝗥𝗨𝘀, 𝗮𝗻𝗱 𝗥𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹-𝗔𝘂𝗴𝗺𝗲𝗻𝘁𝗲𝗱 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻 (𝗥𝗔𝗚). • Provided hands-on training in AI infrastructure and deployment, including 𝐅𝐀𝐈𝐒𝐒, 𝐂𝐡𝐫𝐨𝐦𝐚𝐃𝐁, 𝐋𝐚𝐧𝐜𝐞𝐃𝐁, 𝐀𝐖𝐒 𝐒𝐚𝐠𝐞𝐌𝐚𝐤𝐞𝐫, 𝐀𝐖𝐒 𝐁𝐞𝐝𝐫𝐨𝐜𝐤, 𝐀𝐳𝐮𝐫𝐞, 𝐇𝐮𝐠𝐠𝐢𝐧𝐠 𝐅𝐚𝐜𝐞 𝐒𝐩𝐚𝐜𝐞𝐬, 𝐚𝐧𝐝 𝐍𝐕𝐈𝐃𝐈𝐀 𝐍𝐈𝐌. • Led practical projects focused on deploying and optimizing AI models, connecting generative AI concepts with real-world production use cases.

Education

Bachelor of Technology - BTech, Computer Science(Honours degree), kl university

XII, MPC, Narayana Junior College , Nallakunta

Public links

github: https://github.com/yaswanth169

leetcode: https://leetcode.com/yaswanth169