Being a software engineer in the era of AI - ET CIO
Summary
With clear purpose below two tenets are crucial in this combined mode.⦁ It is important for engineers to ask pertinent questions such as why, what, and how, and to think critically using anti-patterns like why not, what if, and how about.⦁ Once the problem space has been narrowed down, utilize a Generative AI tool with iterative prompting. Refining the prompts with nuances, it is possible to narrow down and decide on the best possible solution, including detailed steps and supporting evidence with pros and cons.However, due to the probabilistic nature of these models, responses may not always align, and applications thus will need to be cross validated. Additionally, working programmatically with APIs and implementing necessary pre and post processing will be crucial to solidify requirement, design, and solutions for high effectiveness.The next level of skills involves building contextual embeddings with vector representations to finetune as a pre-processing chain, which then feeds into LLMs to generate more relevant outcomes as part of design and implementation development. Cross validating outcomes and feeding back the learnings is important for improving accuracy.Finally, deploying automation skills such as MLOps/LLMOps adds on to the current DevOps principles. As engineers continue to upskill in this new age, two key aspects need to be kept in mind.⦁ It is crucial to exercise caution and prevent AI assistance from turning into an addiction, as this will be essential in fostering and elevating human creativity.⦁ With advanced AI agents, it is imperative to continue nurturing curiosity and develop sharp problem-solving skills with HI, complemented by the element of human serendipity.This approach is vital for creating successful products and platforms, as well as advancing ones career in software engineering.